University of Southampton Research Repository ePrints Soton20The%20Role... · 2017. 7. 18. ·...
Transcript of University of Southampton Research Repository ePrints Soton20The%20Role... · 2017. 7. 18. ·...
University of Southampton Research Repository
ePrints Soton
Copyright © and Moral Rights for this thesis are retained by the author and/or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder/s. The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders.
When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given e.g.
AUTHOR (year of submission) "Full thesis title", University of Southampton, name of the University School or Department, PhD Thesis, pagination
http://eprints.soton.ac.uk
UNIVERSITY OF SOUTHAMPTON
FACULTY OF MEDICINE
Academic Unit of Clinical and Experimental Sciences
The Role Of T Cell Subsets In The Airways In Asthma
by
Dr Timothy Stopford Christopher Hinks M.A. (Cantab), B.M.B.Ch., M.R.C.P. (U.K.)
Thesis for the degree of Doctor of Philosophy
January 2013
i
ABSTRACT
UNIVERSITY OF SOUTHAMPTON
ABSTRACT
FACULTY OF MEDICINE
Doctor of Philosophy
Respiratory Medicine and Immunology
THE ROLE OF T CELL SUBSETS IN THE AIRWAYS IN ASTHMA
By Timothy Stopford Christopher Hinks
T-cells are key orchestrators of airways inflammation, but the relative roles of different human T-
cell subsets remain unclear. The aim of my PhD was to carry out a detailed investigation of T
cell phenotypes in asthma in relation to severity and virus-induced exacerbations, with particular
focus on interleukin-17 and TH17 cells, and the recently described mucosal associated invariant
T (MAIT) cells, to improve characterisation of severe asthma versus milder forms of asthma.
A role for interleukin-17 secreting TH17 cells in asthma has been suggested by several
groups. I used clinical and physiologic phenotyping to compare T-cell subsets in health and a
spectrum of different asthma severities. Samples obtained via sputum induction, phlebotomy,
and bronchoscopy were phenotyped using 9-colour flow-cytometry/sorting, RT-qPCR and
multiplex ELISA. The results of my thesis confirm the pre-eminence of TH2 cells in asthma and
provide further evidence of a deficiency of bronchoalveolar Treg in severe asthma, as well as
new evidence of a role for CD8+ Tc2 cells in eosinophilic disease. Conversely, the data do not
indicate a significant role for TH17 or γδ-17 cells in asthma.
Mucosal immunity is intrinsically linked to the associated commensal or pathogenic microbes.
In an exploratory study of these interactions I employed deep-sequencing to characterise the
whole microbial and viral metagenome of the airways in asthma and health.
MAIT cells are novel innate-like T-cells which express an invariant TCRα chain and recognise
the highly-conserved restriction molecule MR1. I observed a selective deficiency of MAIT cells
in asthma, which was not related to age, but exacerbated by systemic corticosteroids and
subject to seasonal variation, indicating their possible regulation by vitamin D. I established
MAIT cell-lines and observed heterogeneity of cytokine expression profiles. These findings open
exciting new avenues for research in this emerging area of T cell biology.
ii
iii
Contents
Introduction ................................................................................................................................... 1
Asthma: an overview ................................................................................................................. 2
Definitions of asthma ............................................................................................................. 2
Asthma a global epidemic ...................................................................................................... 2
A historical perspective .......................................................................................................... 3
Asthma heterogeneity and endotypes ................................................................................... 3
Severe asthma ....................................................................................................................... 4
Asthma exacerbations ........................................................................................................... 5
The pathogenesis of asthma ..................................................................................................... 5
Mast cells ............................................................................................................................... 5
Eosinophils ............................................................................................................................. 6
Basophils ............................................................................................................................... 6
Neutrophils ............................................................................................................................. 7
Macrophages ......................................................................................................................... 7
Inflammatory mediators ......................................................................................................... 8
Innate responses ................................................................................................................... 8
Airways remodelling in asthma .............................................................................................. 9
T lymphocytes (T cells) .............................................................................................................. 9
The importance of T cells in asthma .................................................................................... 10
Interleukin-17 ....................................................................................................................... 11
The T helper 17 subset ........................................................................................................ 12
Interleukin-17, TH17 cells and asthma ................................................................................. 13
IL-17 and TH17 cells in murine models of allergic airways disease ..................................... 15
Regulatory T cells ................................................................................................................ 15
Regulatory T cells in asthma ............................................................................................... 16
CD8+ T cells and asthma .................................................................................................... 17
Mucosal Associated Invariant T (MAIT) cells .......................................................................... 18
Innate-like lymphocytes ....................................................................................................... 18
Mucosal associated invariant T cells ................................................................................... 19
CD161 .................................................................................................................................. 20
MAIT cell restriction ............................................................................................................. 20
MAIT cell ligands .................................................................................................................. 20
MAIT cell development ........................................................................................................ 21
MAIT Cell function ................................................................................................................ 21
MAIT cells in human disease ............................................................................................... 22
MAIT cells and the lung ....................................................................................................... 22
Vitamin D ............................................................................................................................. 23
iv
The Microbiome ....................................................................................................................... 23
The lung microbiome ............................................................................................................ 24
The lung microbiome in cystic fibrosis (CF) ......................................................................... 24
The microbiome in chronic obstructive pulmonary disease (COPD) ................................... 25
The lung microbiome in asthma ........................................................................................... 25
Objectives ................................................................................................................................ 26
Aim 1 .................................................................................................................................... 26
Aim 2 .................................................................................................................................... 27
Aim 3. ................................................................................................................................... 27
Aim 4 .................................................................................................................................... 28
Materials and methods ................................................................................................................ 29
Study design ............................................................................................................................ 30
Clinical measurements ............................................................................................................. 33
Peak flow .............................................................................................................................. 33
Spirometry and reversibility .................................................................................................. 33
Home monitoring .................................................................................................................. 33
TLCO .................................................................................................................................... 33
Exhaled nitric oxide .............................................................................................................. 34
Methacholine challenge testing of airway hyper-responsiveness ........................................ 34
Skin prick allergen testing .................................................................................................... 34
Study populations ..................................................................................................................... 35
Cross sectional study (Aim 1) ............................................................................................... 35
Additional older healthy controls .......................................................................................... 36
Longitudinal study (Aim 2a) .................................................................................................. 36
Clinical classification ................................................................................................................ 36
Asthma control questionnaire ............................................................................................... 36
Phlebotomy .............................................................................................................................. 41
Serum ................................................................................................................................... 41
Full blood count .................................................................................................................... 41
Peripheral blood mononuclear cell preparation .................................................................... 41
Cell preparation tubes .......................................................................................................... 41
Nasal lavage ............................................................................................................................ 41
Sputum induction ..................................................................................................................... 42
Sputum induction protocol .................................................................................................... 42
Sputum processing ............................................................................................................... 42
Preparation of cytospins ....................................................................................................... 42
Definitions of inflammatory subtypes .................................................................................... 43
v
Bronchoscopy .......................................................................................................................... 43
Bronchoscopic technique ..................................................................................................... 43
Processing of BAL ............................................................................................................... 45
Processing of bronchial biopsies ......................................................................................... 45
Collagenase digestion of biopsies ....................................................................................... 46
Culture media .......................................................................................................................... 48
RPMI .................................................................................................................................... 48
Complete serum free medium (AIM V) ................................................................................ 48
RN10 culture medium with 10% human serum ................................................................... 48
T cell growth medium ........................................................................................................... 48
T cell sorting medium ........................................................................................................... 48
Magnetic-activated cell sorting (MACS) Buffer .................................................................... 49
Cryopreservation of cells ......................................................................................................... 49
Thawing cryopreserved cells ............................................................................................... 50
Enzyme linked immunosorbent assay (ELISA) ....................................................................... 50
Measurement of total immunoglobulin E (IgE) .................................................................... 50
Measurement of IL-17 .......................................................................................................... 50
Meso-Scale Discovery platform ........................................................................................... 51
RNA extraction and quantitation .............................................................................................. 52
TRIzol ................................................................................................................................... 52
Nanoprep ............................................................................................................................. 52
Nucleic acid quantitation ...................................................................................................... 53
Reverse transcription and polymerase chain reaction ............................................................ 53
Reverse transcription with SuperScriptTM III RT kit .............................................................. 53
Reverse transcription with Precision nanoScriptTM RT kit ................................................... 53
Polymerase chain reaction (PCR) ....................................................................................... 54
Gel electrophoresis .............................................................................................................. 55
Quantitative PCR ................................................................................................................. 55
Flow cytometry ........................................................................................................................ 55
Surface staining for MAIT cells ............................................................................................ 55
Intracellular cytokine staining ............................................................................................... 57
Cell sorting and data acquisition .......................................................................................... 58
Gating strategy for MAIT cells ............................................................................................. 58
Gating Strategy for T helper cells ........................................................................................ 60
Controls for flow cytometry .................................................................................................. 60
Cloning of MAIT cells ............................................................................................................... 65
Definition of T helper cells for flow cytometry .......................................................................... 65
The problem of CD4 co-receptor downregulation during ex-vivo stimulation ...................... 65
vi
Results .................................................................................................................................. 66
Conclusion ............................................................................................................................ 70
Definitions of Treg for flow cytometry ....................................................................................... 70
Problems with existing markers ............................................................................................ 70
Frequencies of Treg ............................................................................................................. 71
Data analysis to define set-point of FOXP3+ gate ............................................................... 73
Results .................................................................................................................................. 73
Conclusion ............................................................................................................................ 73
Comparison of fresh versus cryopreserved PBMC .................................................................. 73
The need for cryopreservation ............................................................................................. 73
Concerns regarding cryopreservation .................................................................................. 74
Method .................................................................................................................................. 74
Results .................................................................................................................................. 74
Conclusion ............................................................................................................................ 77
Choice of Golgi blocking agent for cryopreserved samples .................................................... 78
Results .................................................................................................................................. 79
Conclusion ............................................................................................................................ 80
Selection and titration of antibodies ......................................................................................... 81
Determination of optimum period of stimulation for MAIT cell intracellular cytokine secretion 84
Methods ................................................................................................................................ 84
Results .................................................................................................................................. 84
Conclusions .......................................................................................................................... 87
Optimisation and validation of RNA extraction method ........................................................... 88
Method .................................................................................................................................. 88
Results .................................................................................................................................. 88
Conclusion ............................................................................................................................ 90
Deep sequencing of the metagenome ..................................................................................... 90
Microarray ................................................................................................................................ 90
Statistical Analysis ................................................................................................................... 91
Data elaboration and preparation for analysis ..................................................................... 91
Cross sectional study ........................................................................................................... 91
Exploratory Analyses of Relationships between variables ................................................... 91
CD4+ T cell phenotypes in asthma ............................................................................................. 95
Introduction .............................................................................................................................. 96
Results and comments ............................................................................................................ 96
Study population ................................................................................................................... 96
Measurement of IL-17 protein by enzyme-linked immunosorbent assay (ELISA) ............... 98
vii
Figure 3.1 ELISA standard curves ....................................................................................... 99
Measurement of serum IgE ............................................................................................... 100
Detection of cytokines by electrochemiluminescence (MSD) ............................................ 100
Cytokines measured by MSD in serum ............................................................................. 103
Cytokines measured by MSD in BAL ................................................................................. 105
Cytokines measured by MSD in sputum ............................................................................ 113
Measurement of IL-17 in airway macrophages by RT-qPCR ............................................ 114
Cytometry of major CD4+ T cell subsets in asthma .............................................................. 116
Evidence of increased TH2 cell inflammation, but no differences in TH17 frequencies in
asthma ............................................................................................................................... 116
Peripheral TH2 responses correlate with atopy and with BAL TH2 cytokines .................... 118
Distinct tissue localisation of different T cell subsets ......................................................... 120
No evidence for a significant role of TCR+ IL-17+ T cells in human asthma ................. 121
No evidence for IL-17 producing TH2 cells in human asthma ............................................ 124
Analysis of CD4+ T cells according to inflammatory subtype ............................................ 125
Cluster analysis to explore relationships between variables ............................................. 125
Discussion ............................................................................................................................. 127
The fundamental role of TH2 inflammation in asthma ........................................................ 127
Evidence for a deficiency of regulatory T cells .................................................................. 127
The uncertain significance of interleukin-17 ...................................................................... 129
Relegating TH17 cells ......................................................................................................... 131
T-cells ............................................................................................................................. 132
CD8+ T cells in asthma ............................................................................................................. 134
Introduction ............................................................................................................................ 135
Results and comments .......................................................................................................... 135
Study population ................................................................................................................ 135
Definitions of T cell subsets ............................................................................................... 135
Type 2 cytokine-secreting cytotoxic T cell frequencies are increased in asthma in PBMC
and BAL, and correlate with disease severity .................................................................... 135
Type I cytokine-secreting cytotoxic T cell are increased only in BAL, in mild asthma. ..... 138
Frequencies of IL-17 secreting cytotoxic T cells are not associated with asthma ............. 144
Clinical correlations with Tc2 cell frequencies ................................................................... 144
Type 2 cytokine-secreting cytotoxic T cell frequencies according to inflammatory subtype,
nasal polyposis and history of smoking ............................................................................. 144
Clinical correlates of peripheral blood Tc2 cell frequencies .............................................. 147
Preliminary analysis of the T cell transciptome is supportive of a role for CD8+ T cells in
asthma ............................................................................................................................... 147
Discussion ............................................................................................................................. 151
What is known of a link between CD8+ cells and eosinophils in asthma? ........................ 151
viii
The role of Tc1 cells in asthma .......................................................................................... 153
Conclusion .......................................................................................................................... 154
MAIT cells – new players in asthma .......................................................................................... 155
Introduction ............................................................................................................................ 156
Results and comments .......................................................................................................... 156
Study population ................................................................................................................. 156
Analysis of MAIT cells in human asthma ............................................................................... 158
MAIT cells are deficient in human asthma and correlate with disease severity ................. 158
MAIT cell frequencies are not related to age ..................................................................... 160
Clinical correlations with MAIT cell frequencies ................................................................. 163
Modulation of MAIT cell frequencies by corticosteroids ......................................................... 164
Inhaled corticosteroids ....................................................................................................... 166
Oral corticosteroids............................................................................................................. 166
Seasonal variation in MAIT cell frequencies .......................................................................... 168
Rationale for investigating seasonal variation .................................................................... 168
Seasonal variations in MAIT cell frequencies .................................................................... 168
Development and characterisation of MAIT cell clones ......................................................... 171
Cloning technique ............................................................................................................... 171
Confirmation of MAIT clones by PCR ................................................................................. 174
Clone phenotype ................................................................................................................ 174
Discussion .............................................................................................................................. 176
Conclusions ........................................................................................................................ 178
Deep sequencing of the airway microbiome ............................................................................. 179
Introduction ............................................................................................................................ 180
Results and comments .......................................................................................................... 182
Participants ......................................................................................................................... 182
Results section I ..................................................................................................................... 183
a) Bacterial species in BAL from the pilot study ................................................................. 183
b) Bacterial species in BAL from the main study ................................................................ 185
c) Bacterial species in sputum from the pilot study ............................................................ 188
d) Bacterial species in sputum from the main study ........................................................... 190
Summary ............................................................................................................................ 190
Results section II .................................................................................................................... 190
Viral species in sputum and BAL ........................................................................................ 190
Discussion .............................................................................................................................. 192
Microbial modulation of the respiratory immune system .................................................... 192
ix
Low bacterial frequencies argue against a significant airways microbiome ...................... 193
Unique contributions from this thesis ................................................................................. 194
No evidence of chronic respiratory viral infection in asthma ............................................. 195
Conclusion ............................................................................................................................. 197
Future work ........................................................................................................................ 197
T cell phenotypes during natural cold-induced asthma exacerbations ..................................... 199
Introduction ............................................................................................................................ 200
The nature of asthma exacerbations ................................................................................. 200
The immune response to rhinovirus .................................................................................. 200
Study design .......................................................................................................................... 202
Interferon beta study .......................................................................................................... 202
Immunological samples ..................................................................................................... 204
Study populations .................................................................................................................. 205
Pilot RV16 challenge study ................................................................................................ 205
Interferon-beta study longitudinal cohorts .......................................................................... 205
Results I Analysis of pilot data from RV challenge cohort ..................................................... 208
Induction of IL-17 mRNA in sputum during experimental RV infection ............................. 208
Results II Analysis of fresh samples from longitudinal cohort ............................................... 209
T cell frequencies in peripheral blood and sputum during acute viral infection ................. 209
The effect of IFN-β1α on T cell frequencies in blood and sputum ..................................... 211
Results III Analysis of cryopreserved PBMC samples from longitudinal cohort .................... 212
TH17 cell frequencies in peripheral blood are elevated during treatment with inhaled rhIFN-
β1α ..................................................................................................................................... 217
TH17 cell frequencies in peripheral blood are according to whether subjects suffer an
asthma exacerbation .......................................................................................................... 217
Discussion ............................................................................................................................. 219
Respiratory virus infections are not associated with a TH17 response .............................. 219
Administration of inhaled rhIFN-β1α is associated with increased TH17 frequencies in
peripheral blood ................................................................................................................. 220
Conclusion ............................................................................................................................. 222
Discussion ................................................................................................................................. 223
The fundamental role of TH2 inflammation in asthma ........................................................ 224
The history of interleukin-17 and TH17 cells in asthma highlights research pitfalls ........... 224
A renewed interest in CD8+ T cells in asthma is warranted .............................................. 226
The need for the application of deep sequencing to the study of asthma ......................... 226
MAIT cells as a priority for future research ........................................................................ 228
Future work ............................................................................................................................ 228
x
Deep sequencing of the microbiome during exacerbations ............................................... 228
An integrated systems biology approach to the analysis of transcriptomic data obtained
from microarray of epithelial cells and pure T cell populations .......................................... 229
A characterisation of the function of MAIT cells in human lung diseases .......................... 229
References ................................................................................................................................ 231
xi
xii
List of figures
Figure Page
1.1 Figure IL-17 and TH17 cells in the immune response 14
2.1 Cross sectional study flow diagram 31
2.2 Longitudinal study flow diagram 32
2.3 BTS treatment algorithm 37
2.4 Sample processing 45
2.5 Cleavage of CD4 by collagenase dispersion 47
2.6 Gating strategy for MAIT cells 59
2.7 Gating strategy T helper cells 61
2.8 Controls for cytometry 62
2.9 Comparison of isotypes and unstimulated cells 64
2.10 Changes in CD4 and CD8 populations with stimulation 67
2.11 CD4 receptor down regulation 69
2.12 Setting of regulatory T cell gates 72
2.13 Intracellular cytokine staining in fresh and cryopreserved cells 75
2.14 The effect of cryopreservation on measurement of specific T cell subsets 77
2.15 Comparison of two inhibitors of Golgi function 80
2.16 Titration of antibodies 82
2.17 Determination of optimum period of stimulation for MAIT cell intracellular 85
2.18 Comparison of RT-qPCR on fixed and unfixed T cells 89
3.1 ELISA standard curves 99
3.2 Validation of MSD in sputum 101
3.3 Cytokines measured by multiplex ELISA in serum 104
3.4 Cytokines measured by multiplex ELISA in bronchoalveolar lavage 105
3.5 Cytokines measured by multiplex ELISA in sputum 106
3.6 Cytokines measured by multiplex ELISA compared between asthma and
health
107
3.7 Correlates of BAL IL-17 levels 108
3.8 Relationship between BAL IL-17 levels and BAL epithelial cells 109
3.9 Correlates of airway TH2 cytokines 110
3.10 Airway cytokines according to inflammatory phenotype 111
3.11 Airway macrophage expression of IL-17 mRNA 114
3.12 Major CD4+ T cell subsets in asthma and health 115
3.13 Major CD4+ T cell subsets stratified by disease severity 117
3.14 Ratio of TH2:TH1 cells in different tissue compartments 118
xiii
3.15 Correlates of high peripheral blood TH2 frequencies 119
3.16 Compartmentalisation of tissue CD4+ T cells 120
3.17 γδ T cells in asthma 122
3.18 No evidence for TH2/17 cells in humans 123
3.19 CD4+ T cell frequencies stratified by inflammatory cell subtype 124
4.1 Type 2 cytokine-secreting cytotoxic T cell frequencies are increased in asthma
in PBMC and BAL
136
4.2 Type 1 cytokine-secreting cytotoxic T cell frequencies are increased in BAL in
asthma
137
4.3 Type 1 cytokine-secreting cytotoxic T cells are increase in BAL in mild asthma 138
4.4 A comparison of Tc1 and TH1 cells in BAL 139
4.5 Frequencies of IL-17-secreting CD8+ T cell do not differ asthma 140
4.6 Correlations between Tc2 and TH2 cells in tissues 141
4.7 Type 2 cytokine-secreting cytotoxic T cell frequencies correlate with disease
severity in blood
142
4.8 Type 2 cytokine-secreting cytotoxic T cell frequencies according to
inflammatory subtype, nasal polyposis and history of smoking
143
4.9 Ratio of TC1:TC2 T cells according to inflammatory subtype 145
4.10 Clinical correlates of peripheral blood Tc2 cell frequencies 146
4.11 T cell associated networks are down-regulated in severe asthma 148
4.12 Hierarchical clustering of asthma v health in BAL T cells reveals a strong
asthma-associated gene signature
149
4.13 Hierarchical clustering of asthma v health in sputum T cells reveals a strong
asthma-associated gene signature
150
5.1 MAIT cells are deficient in asthma 158
5.2 MAIT cell deficiency correlates with asthma severity 159
5.3 Frequencies of a non-MAIT T cell subset do not differ in asthma 160
5.4 MAIT cell frequencies are not related to age 161
5.5 Clinical correlates of peripheral blood MAIT cell frequencies 162
5.6 MAIT cell frequencies and use of inhaled corticosteroids 164
5.7 MAIT cell and non-MAIT cell frequencies before and after inhaled
corticosteroids
166
5.8 MAIT cell and non-MAIT cell frequencies before and after oral corticosteroids 167
5.9 Annual variation in MAIT cell frequencies 169
5.10 Surface phenotype of MAIT clones 172
xiv
5.11 Confirmation of that MAIT clones express the invariant Vα7.2-Jα33 TCR
rearrangement
173
5.12 Typical intracellular cytokine expression by a stimulated MAIT clone 174
5.13 Heterogeneity of cytokine expression profile of MAIT clones 175
6.1 Proportions of bacterial taxa in each bronchoalveolar lavage sample 186
6.2 Bacterial abundance in bronchoalveolar lavage 187
6.3 Viral taxa in bronchoalveolar lavage samples 191
7.1 Sputum IL-17 mRNA during experimental RV infection 208
7.2 T cell frequencies in peripheral blood and sputum during acute viral infection 210
7.3 T cell frequencies in peripheral blood and sputum stratified by treatment group 212
7.4 T cell frequencies in cryopreserved peripheral blood during an acute viral
infection
214
7.5 T cell frequencies in cryopreserved peripheral blood during an acute viral
infection: showing individual subjects separately
215
7.6 Peripheral blood T cell subsets according to treatment group 216
7.7 Peripheral blood TH17 response according to treatment group 217
7.8 Peripheral blood TH17 response according to whether exacerbated 218
xv
xvi
List of tables
Table Page
2.0 Interpretation of methacholine challenge testing 34
2.1 Levels of asthma control, GINA 39
2.2 Definitions of asthma severity used in this project 40
2.2.1 Definitions of inflammatory subtypes 43
2.3 Oligonucleotide primers used for PCR 54
2.4 Antibodies and fluorochromes used for surface staining 56
2.5 Isotype controls 56
2.6 Stimulation times for each tissue 57
2.6.1 Antibodies and fluorochromes used for intracellular staining 58
2.7 The effect of stimulation on relative T cell populations in different tissues. 68
2.8 Results of Treg set-point analysis. PBMC T cell frequencies in n=15 healthy 73
2.9 Median frequencies of T cell subsets assessed by intracellular staining with or 76
3.1 Demographic and clinical characteristics of cross sectional cohort for CD4+
and CD8+ T cell analysis
97
3.2 Percentage of cytokine measured in DTE/diluent 1:1 mix compared with that in
proprietary diluent alone across the lower dynamic range
102
3.3 Average spiking recovery from sputum using 10 pg/ml spikes 102
3.4 Average spiking recovery from BAL using 10 pg/ml spikes 102
3.5 Effective limits of detection for cytokines measurement by MSD for each tissue 103
3.6 Principle component analysis of data from the cross sectional study 126
4.1 Numbers of successful microarrays performed and passing quality data quality
control
147
5.1 Clinical characteristics of MAIT cell study population 157
6.1 Bacterial OTU identified from BAL in the pilot dataset 183
6.2 Bacterial OTU identified by more than one read from sputum samples
collected during acute viral upper respiratory tract infections
189
7.1 Study schedule for longitudinal study 203
7.2 Clinical characteristics of the longitudinal cohort (fresh samples) 206
7.3 Clinical characteristics of the longitudinal cohort (cryopreserved) 207
7.4 Rates of successful sputum inductions during longitudinal study 211
xvii
xviii
List of accompanying materials
Staples, K. J., T. S. Hinks, et al. (2012). "Phenotypic characterization of lung macrophages in
asthmatic patients: Overexpression of CCL17." J Allergy Clin Immunol 130(6): 1404-1412
e1407.
xix
xx
DECLARATION OF AUTHORSHIP
I, Dr Timothy Stopford Christopher Hinks
declare that the thesis entitled
‘The Role Of T Cell Subsets In The Airways In Asthma’
and the work presented in the thesis are both my own, and have been generated by me as the
result of my own original research. I confirm that:
this work was done wholly or mainly while in candidature for a research degree at this
University;
where any part of this thesis has previously been submitted for a degree or any other
qualification at this University or any other institution, this has been clearly stated;
where I have consulted the published work of others, this is always clearly attributed;
where I have quoted from the work of others, the source is always given. With the exception
of such quotations, this thesis is entirely my own work;
I have acknowledged all main sources of help;
where the thesis is based on work done by myself jointly with others, I have made clear
exactly what was done by others and what I have contributed myself;
none of this work has been published before submission
Signed: ………………………………………………………………………..
Date:…………………………………………………………………………….
xxi
xxii
Acknowledgements
I wish to thank my supervisors Prof Ratko Djukanović and Prof Stephan Gadola for their
inspiration, dedicated supervision, and invaluable guidance. I am also grateful for the support
and advice of my colleagues within the Djukanović Inflammatory Cell Biology Group – Dr Karl
Staples, Dr Ben Nicholas, Dr Asha Ganesan, Pam Sunder – and of scientists within the
Southampton Respiratory Biomedical Research Unit: Kerry Gove for assistance with pulmonary
physiology, Richard Jewell and Carolann McGuire for assistance with flow cytometry, and Dr
Laurie Lau and Clair Barber. I am graeful for the help of Dr Salah Mansour in the Gadola group
for help with T cell cloning.
I am indebted to the clinical team of the Southampton Respiratory Biomedical Research Unit for
their bronchoscopy nursing support: Caroline Smith, Martina Brown, Lesley-Ann Castle, Shuna
Egerton, Lisa Hewitt, Louise Hoile, Dr Kathleen Holding, Malcolm North, Sandy Pink, Kerry
Thorpe and to Dr Paddy Dennison, Dr Tom Havelock, Dr Hitasha Rupani for help with clinics. I
am grateful to Jon Ward for staining and analysis of cytospins and to Prof Peter Howarth for his
collaboration in providing access to the Wessex Severe Asthma cohort. I am grateful for
statistical support from Dr Borislav Dimitrov in the conduct of the cluster analysis and
longitudinal analysis of variance.
Exacerbation samples were provided through a collaboration with the research team at
Synairgen Research Ltd, Southampton. Samples of peripheral blood and sputum were collected
at Southampton by the team, of which I was an unpaid member. Synairgen scientists also
isolated and cryopreserved peripheral blood cells for me. The research team comprised Dr
Peter Adura, Dr Valia Kehagia, Dr Florian Gahleitner, Paul Rucki, Joanna Samways, Sarah
Bavington, Jody Brookes, Lara Balls, Kate Mutendera, Christine Boxall, Rona Beegan, Thelma
Deacon, Sarah Dudley, Jayne O'Hara, James Roberts, Kerry Lunn, Lauren Cracknell, Sarah
Hrebien, Dr Cathy Xiao, Victoria Tear and Dr Phil Monk.
Analysis of serum vitamin D levels was performed at University Hospitals Southampton which
was arranged by Prof Alan Jackson and Dr Steve Wootton, at the Southampton Biomedical
Research Centre in Nutrition.
Samples I collected for microbiological analysis were sequenced and analysed by Prof Larissa
Thackray, Dr Lindsay Droit, Dr Scott Handley, Dr Dave Wang and Dr Guoyan Zhao in the
laboratory of my collaborator the laboratory of Prof Herbert ‘Skip’ Virgin VI, at Washington
University School of Medicine, St Louis.
xxiii
Microarray analysis was performed and analysed by Dr Daniel Horowitz, Dr Fred Baribaud, Dr
Anuk Das and Dr Anthony Rowe of Janssen Research & Development, Springhouse,
Pennsylvania.
I will always be grateful for those who have previously inspired and encouraged me to pursue
academic medicine, in particular Dr James CD Hickson, Prof Ajit Lalvani and Dr Gerrard
Phillips. I would have achieved none of this without the support, love and patient forbearance of
my wife Naomi.
I am thankful for the generous participation of all the research volunteers involved.
This work was made possible by my personal award of a Clinical Research Training Fellowship
from the Wellcome Trust. I also acknowledge the support of the National Institute for Health
Research, through the Primary Care Research Network, and through an Academic Clinical
Fellowship.
This work is dedicated to the memory of
Dr Justus Kenneth Landquist, Dr F Christopher Maddox and Dr Robert J Davies
soli Deo gloria
xxiv
xxv
Definitions and abbreviations
ACQ Asthma control questionnaire
AHR Airway hyper-responsiveness
AIM V® Adoptive Immunotherapy Media V®
APC Allophycocyanin
APS Airway provocation system
ATP Adenosine triphosphate
ATS American Thoracic Society
β2M β-2microglobulin
BAL Broncho-alveolar lavage
BCG Bacillus Calmette-Guérin
BDP Beclomethasone dipropionate
BHR Bronchial hyper-responsiveness
BLAST Basic Local Alignment Search Tool
β-ME 2-mercaptoethanol (β-mercaptoethanol)
BSA Bovine serum albumin
BTS British Thoracic Society
CD Complementarity determinant
COPD Chronic obstructive pulmonary disease
cDNA Complementary DNA
CF Cystic fibrosis
CRTH2 Chemoattractant receptor-homologous molecule expressed on Th2 cells
CCL Chemokine (C-C motif) ligand
CXCL Chemokine (C-X-C motif) ligand
Cy Cyanine
DC Dendritic cell
ddH20 Double distilled water treated with DEPC
DEPC Diethylpyrocarbonate
1,25(OH)2D3 1,25-dihydroxy vitamin D(3)
DMSO Dimethyl sulphoxide
DNA Deoxyribonucleic acid
DNase Deoxyribonuclease
dNTP Deoxyribonucleotide triphosphate
DTE Dithioerythritol
DTT Dithiothreitol
EAE Experimental autoimmune encephalomyelitis
xxvi
EDTA Ethylenediamine tetraacetic acid
ELISA Enzyme-linked immunosorbent assay
eNO Exhaled nitric oxide
ERS European Respiratory Society
FcγR Fragment crystallisable gamma receptor
FCS Foetal calf serum
FENO Fractional exhaled nitric oxide (see eNO)
FER Forced expiratory ratio (FEV1/FVC)
FEV1 Forced expiratory volume in 1 second
FISH Fluorescent in-situ hybridisation
FITC Fluorescein isothiocyanate
FOXP3 Forkhead box P3
FSC Forward scatter
FVC Forced vital capacity
GAPDH Glyceraldehyde 3-phosphate dehydrogenase
GI Gastrointestinal
GINA Global Initiative for Asthma
GRO-α GRO1 oncogene-α (CXCL1)
HBSS Hank’s balanced salt solution
HEPES N-[2-hydroxyethyl] piperazine-N’-[2-ethanesulfonic acid]
HRCT High resolution computed tomography
HRP Horse radish peroxidase
HSA Human serum albumin
ICAM Intercellular adhesion molecule
IFN Interferon
IFNAR Interferon-α/β receptor
Ig Immunoglobulin
IL Interleukin
IP10 IFN-γ-inducible protein 10 (CXCL10)
iNKT Invariant natural killer T cell
ITAC Interferon-inducible T-cell alpha chemoattractant (CXCL11, IP9)
IU International units
MACS Magnetic-activated cell sorting
MADscore Median absolute deviation score
MAIT Mucosal associated invariant T cell
MHC Major histocompatibility complex
mL Millilitres
xxvii
μL Microlitres
mM Millimolar
μM Micromolar
mRNA Messenger ribonucleic acid
MS Multiple sclerosis
MTB Mycobacterium tuberculosis
MxA Myxoma resistance gene A
NIBSC National Institute for Biological Standards and Control
OTU Operational taxonomic unit
PBMC Peripheral blood mononuclear cells
PBS Phosphate buffered saline
PCA Principal component analysis
PCR Polymerase chain reaction
PC20 Provocative concentration of methacholine causing a 20% drop in FEV1
(PC20FEV1).
PD20 Provocative dose of methacholine causing a 20% drop in FEV1
PE R-phycoerythrin
PEFR Peak expiratory flow rate
PerCP Peridinin chlorophyll-protein
pH Negative log of hydrogen ion concentration
PHA Phytohaemagglutinin
PMA Phorbol 12-myristate 13-acetate
qPCR Quantitative polymerase chain reaction
RANTES Regulated and normal T cell expressed and secreted (CCL5)
RMA Robust multi-array average
RNA Ribonucleic acid
RNase Ribonculease
ROR Retinoic acid-related orphan nuclear hormone receptor
RPM Revolutions per minute
RPMI Roswell Park Memorial Institute medium
RSV Respiratory syncytial virus
RT Room temperature or reverse transcription
RT-PCR Reverse transcription-polymerase chain reaction
RV Rhinovirus
SCFA Short chain fatty acid
SSC Side scatter
Src Sarcoma
xxviii
Taq Thermus aquaticus
TBE Tris/Borate/EDTA
Tc Cytotoxic T cell (CD8+ T lymphocyte)
TCID50 Tissue culture infective dose 50
TCR T cell receptor
TDI Toluene diisocyanate
TGFβ Transforming growth factor β
TH T helper cell (CD4+ T lymphocyte)
TLCO Transfer Factor of Lung Carbon monoxide
TLR Toll-like receptor
TMB Tetramethyl-benzidine
TNF Tumour necrosis factor
TREG Regulatory T cell
T-RFLP Terminal restriction fragment length polymorphism
Tris Tris(hydroxymethyl) aminoethane
TTMV Torque Teno Mini Virus
Tween 20 Polyoxythylenesorbitan monolaurate
UBC Ubiquitin C
WHO World Health Organisation
XIAP X-linked inhibitor of apoptosis
YWHAZ Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein,
zeta polypeptide
xxix
Timothy SC Hinks 1. Introduction
1
CHAPTER 1
Introduction
Magna opera Domini esquisira in ornnes coluntares ejrts 1
1 Inscription carved into the great oak doors of the Cavendish Laboratory, Free School
Lane in Cambridge at the request of Prof James Clerk Maxwell FRS FRSE (1831-1879).
Known as the research scientist’s text, this was replicated in 1973 over the entrance to
the New Cavendish Laboratories. It may be translated ‘Great are the works of the Lord;
they are pondered by all who delight in them.’ [Psalm 111:2, NIV]
Timothy SC Hinks 1. Introduction
2
The aim of my thesis is to report on a detailed study of T cell phenotypes in asthma in relation to
asthma severity and virus-induced asthma exacerbations. My work focuses particularly on two
novel T cell subsets: the T helper 17 cell (TH17) and the mucosal associated invariant T (MAIT)
cell. Therefore my introduction will provide a brief general review of the nature of asthma and
the role of various inflammatory cell types in its pathogenesis, before discussing in much more
detail what is known about TH17 cells and the related but functionally antagonistic regulatory T
(Treg) cell subset. I will then provide a review of the emerging literature regarding MAIT cells.
The activation of these innate and adaptive responses within a mucosal immune system may be
related intrinsically to the associated microbial flora which I have therefore also attempted to
characterise, and so I will review current knowledge of the nature of the airway microflora. I will
conclude this introduction by outlining the specific hypothesis I have undertaken to test.
Asthma: an overview
Definitions of asthma
The Global Initiative for Asthma (GINA) defines asthma as ‘a chronic inflammatory disorder of
the airways in which many cells and cellular elements play a role. The chronic inflammation is
associated with airway hyper-responsiveness that leads to recurrent episodes of wheezing,
breathlessness, chest tightness, and coughing, particularly at night or in the early morning.
These episodes are usually associated with widespread, but variable, airflow obstruction within
the lung that is often reversible either spontaneously or with treatment’ ((GINA) 2010).
Asthma a global epidemic
Asthma affects 5 million people in the UK (Holgate 2004) and 150-300 million worldwide and the
prevalence is increasing (Cookson 1999; Adcock, Caramori et al. 2008; Anderson 2008).
Asthma was uncommon at the start of the 20th century, but in developed countries prevalence
particularly of atopic asthma (Upton, McConnachie et al. 2000) has risen dramatically, roughly
doubling over the last 20-30 years (Fleming and Crombie 1987; Aberg, Hesselmar et al. 1995)
and becoming a true global epidemic (Cookson 1999). Whilst some of the increase may be
spurious and due to increased diagnosis (Rona, Chinn et al. 1995), the scale of the
epidemiological changes and observations such as tenfold regional differences in prevalence
imply some major change or changes in environmental factors (Cookson 1999). Several
epidemiological observations are well recognised: asthma and allergic diseases are less
common in non-westernized environments, in rural environments, amongst children of livestock
farmers, amongst younger siblings, and in households with dogs as pets (Strachan 1989;
Cookson 1999). A wide variety of potentially causative environmental factors have been
proposed, including early life exposure to house dust mite (Sporik, Holgate et al. 1990), or to
infections – ‘the hygiene hypothesis’ (Strachan 1989; Lewis, Butland et al. 1996; Strachan
2000) - to changes in gastrointestinal microbiome (von Mutius, Fritzsch et al. 1992),
Timothy SC Hinks 1. Introduction
3
breastfeeding practices (Wills-Karp, Brandt et al. 2004), exposure to paracetamol (Beasley,
Clayton et al. 2008), chlorine (Bernard, Carbonnelle et al. 2003) and diesel fumes (Diaz-
Sanchez, Proietti et al. 2003), or nutritional factors such as vitamin D and obesity (Chinn 2003;
Weiss and Litonjua 2011). As yet no data have provided a decisive explanation for this ongoing
epidemic (Cookson 1999).
A historical perspective
Asthma is not a single disease but a spectrum of disorders characterised by airway obstruction
that varies spontaneously and with treatment (Barnes, Djukanovic et al. 2003). The term
‘asthma’ derives from the Greek ἅσθμα – first used in Homer’s Iliad (Homer) - meaning ‘to
exhale with open mouth, to pant’ and has been used in English since around 1600 (Keeney
1964), although the earliest descriptions of asthma perhaps date back to a Chinese medical
textbook c2600BC (Walter and Holtzman 2005). In the first century AD Seneca provided a vivid
personal description of asthma, stressing its sudden onset and periodic nature (Seneca 65-65
AD; Panzani 1988). By the 19th century in his classic work the Dorset born physician Henry
Hyde Salter defined asthma as ‘Paroxysmal dyspnoea of a peculiar character, generally
periodic with intervals of healthy respiration between attacks’ (Hyde 1860; Sakula 1985). This
key element of variability over time has been retained in more modern definitions (Bousquet,
Jeffery et al. 2000; (GINA) 2010) and reflects the close link between asthma and underlying
allergic airways inflammation.
In 1905 von Pirquet and Schick reported the first clinical observations of anaphylactic reactions
in children caused by hypersensitiveness to horse serum (von Pirquet and Schick 1905). The
term allergy was introduced by Pirquet a year later to describe the skin reaction following
subcutaneous injection of tuberculin in sensitised individuals (Von Pirquet 1906). In 1910
Meltzer suggested that asthma was a manifestation of anaphylaxis, prompted by the earlier
studies of Auer, who noted bronchospasm and pulmonary distension in guinea pigs dying of
anaphylactic shock (Meltzer 1910). The term ‘atopy’ was introduced by Coca (1923) to apply to
hypersensitivity mediated by an antigen-antibody mechanism, and in which hereditary
influences may play an important role (Coca and Cooke 1923), and is now understood to be
caused by an exaggerated tendency to mount IgE responses to a wide variety of common
environmental allergens (Holgate 1999; Murphy, Travers et al. 2008). This tendency is variably
expressed in the distinct but immunologically related conditions of eczema in the skin, allergic
rhinitis in the upper airways and, where exposure to aeroallergens triggers airways
inflammation, as allergic asthma (Holgate 1999).
Asthma heterogeneity and endotypes
Disease heterogeneity has long been described (Rubin and Rubin 1947) but its relevance to
understanding the mechanisms underlying asthma has risen to prominence only lately
(Anderson 2008). In recent years the analysis of large, carefully phenotyped cohorts of
Timothy SC Hinks 1. Introduction
4
asthmatics by the statistical technique of cluster analysis has led to the improved definition of
distinct asthmatic clinical phenotypes (Anderson 2008; Haldar, Pavord et al. 2008; Moore,
Meyers et al. 2010). Haldar et al described two phenotypes of severe refractory asthma both
characterised by discordance between symptoms and eosinophilic airway inflammation, which
they termed early-onset symptom predominant and late-onset inflammation predominant
subsets (Haldar, Pavord et al. 2008). In a larger study of 726 subjects Moore et al. identified five
phenotypic clusters: 1) early onset atopic asthma with normal lung function; 2) early-onset
atopic asthma and preserved lung function with increased medication requirements, 3) older
obese women with late-onset non-atopic asthma, moderate reductions in FEV1 and frequent
oral corticosteroid, and 4) and 5) with severe airflow obstruction and bronchodilator
responsiveness but differing in their ability to attain normal lung function, age of asthma onset,
atopic status and use of oral corticosteroids (Moore, Meyers et al. 2010).
Severe asthma
Arising from these and other studies it is apparent that a subgroup of 5-10% of asthmatics have
severe disease despite anti-inflammatory therapy and airway inflammation characterised by
neutrophilic infiltration (Wenzel, Szefler et al. 1997; Gibson 2007; Adcock, Caramori et al. 2008).
These subjects frequently meet the American Thoracic Society 2000 consensus definition of
severe refractory asthma (2000), which requires at least one major criterion and two minor
criteria are met, the exclusion of other disorders, the treatment of exacerbating factors and
generally good patient compliance:
Major characteristics
• Treatment with continuous or near continuous (≥50% of year) oral corticosteroids
• Need for treatment with high-dose inhaled corticosteroids
Minor characteristics
• Need for additional daily treatment with a controller medication (eg, long-acting β agonist,
theophylline, or leukotriene antagonist)
• Asthma symptoms needing short-acting β agonist use on a daily or near-daily basis
• Persistent airway obstruction (FEV1 <80% predicted, diurnal peak expiratory flow
variability >20%)
• One or more urgent care visits for asthma per year
• Three or more oral steroid bursts per year
• Prompt deterioration with ≤25% reduction in oral or intravenous corticosteroid dose
• Near-fatal asthma event in the past
Wenzel et al investigated severe, steroid-dependent asthmatics bronchoscopically and found
that the severe asthmatics had higher levels of neutrophils in bronchoalveolar lavage (BAL) and
bronchial biopsies than either mild-moderate asthmatics or normal controls, suggesting that
Timothy SC Hinks 1. Introduction
5
neutrophilic airways inflammation may be at least one mechanism for steroid refractory disease
(Wenzel, Szefler et al. 1997).
Asthma exacerbations
The classic variability of asthmatic symptoms is seen most dramatically during acute
exacerbations, which are also associated with steroid-refractory inflammation (Grunberg,
Sharon et al. 2001). Exacerbations are a major cause of morbidity and mortality (Anderson
2008), as well as conferring a substantial financial cost in terms of healthcare expenditure and
lost productivity (Cookson 1999). In severe asthma five risk factors have been identified for
recurrent exacerbations: severe nasal sinus disease, gastro-oesophageal reflux, recurrent
respiratory infections, psychological affective disorders and obstructive sleep apnoea (Anderson
2008). However, irrespective of these predisposing risk factors, it is now well documented that
the direct trigger factors in the vast majority of these exacerbations are viral infections of the
upper respiratory tract (Johnston, Pattemore et al. 1995; Johnston, Pattemore et al. 1996).
Viruses are detected by PCR in approximately 80% of exacerbations(Johnston, Pattemore et al.
1995) and are associated with airway neutrophilia (Wark, Johnston et al. 2002). Individuals with
atopic asthma are not at greater risk of upper airways viral infections than healthy individuals
but suffer from more frequent lower respiratory tract (LRT) infections and have more severe and
longer-lasting LRT symptoms (Corne, Marshall et al. 2002). Studies of epithelial cultures
infected with rhinovirus 16 have demonstrated that the mechanism that explain this
susceptibility is a defect in the production of type I (Wark, Johnston et al. 2005) and type III
interferons (Contoli, Message et al. 2006), leading to a failure of apoptosis that normally
develops as a consequence of virus infection; instead the infected cell undergoes cytolysis
when infected, thereby leading to increased viral replication and dissemination within the lower
airways (Wark, Johnston et al. 2005).
The pathogenesis of asthma
Whilst diverse mechanisms may underlie the collection of diseases which comprise the
syndrome of asthma, common to all are patterns of mucosal inflammation involving activated
inflammatory mast cells, eosinophils and T lymphocytes, and with associated altered responses
of structural cells in the airways, including epithelial cells, fibroblasts, endothelial cells and
smooth muscle cells (Holgate, Lackie et al. 2001; Holgate and Polosa 2006; Holgate 2008). To
help place T lymphocytes in the appropriate immunological context I will review briefly what is
known of these other key cell types in the pathogenesis of asthma.
Mast cells
Mast cells are found throughout the airways especially within the bronchial epithelium and
submucosa (Flint, Leung et al. 1985) but are rare within the lumen. They are key mediators of
type-I hypersensitivity reactions in which inhaled aeroallergens cross-link IgE on the surface of
mast cells causing rapid degranulation. This releases a variety of pro-inflammatory mediators
Timothy SC Hinks 1. Introduction
6
including histamine – which directly causes bronchoconstriction, changes in bronchial arterial
perfusion and microvascular leakage – as well as the mast cell proteases tryptase, chymase,
carboxypeptidase, cathepsin G, elastase, plasminogen activator and matrixmetalloproteinase
(MMP)-9 (Macfarlane, Kon et al. 2000). These mediators can exacerbate bronchoconstriction
via activation of bradykinin. Activated mast cells also synthesise new mediators including
arachidonic acid metabolites such as the leukotrienes which also promote bronchoconstriction
and airways inflammation (Laidlaw and Boyce 2012). It is also recognised that mast cell
produce a variety of cytokines which had previously been attributed to T cells (Bradding,
Roberts et al. 1994).
Eosinophils
Eosinophils have long been associated with asthma by their presence in sputum and the
mucosa, their association with clinical responsiveness to steroids and their abundance in the
airways in post-mortem studies in asthma(Brightling 2011). Sputum eosinophilia (defined as
eosinophils comprising >3% of airway respiratory cells)(Pavord, Brightling et al. 1999; Green,
Brightling et al. 2002) is correlated with bronchial hyper-responsiveness and with steroid
responsive disease. In turn, airway eosinophilia is correlated with measured levels of exhaled
nitric oxide. Eosinophils express the low affinity IgE receptor and are believed to play an
important role in the late-phase reaction to inhaled aeroallergens by an IgE dependent
mechanism(Durham 1998), releasing oxygen free radicals, leukotrienes and Th2 cytokines,
growth factors and MMPs (Wardlaw, Brightling et al. 2000).
Basophils
Basophils are the rarest circulating granulocyte, sharing many functional characteristics with
tissue-resident mast cells, and are generally associated with type 2 immune responses
(Voehringer 2011). They can leave the circulation to reach tissues where they are able to
survive for several weeks. Their role in asthma is the least well defined of all inflammatory cells,
but they are known to strongly secrete IL-4 and IL-13 both of which are implicated in atopic
disease. Like mast cells they express the high affinity IgE receptor FcRI and contain basophilic
granules, which can produce a wide variety of inflammatory mediators including histamine,
platelet-activating factor, leukotriene C4, IL-4 and IL-13. They differ from mast cells in their
relative inability to proliferate and perform phagocytosis, their lower responsiveness to
complement, and their greater steroid responsiveness (Djukanovic, Wilson et al. 1992).
Accumulations of basophils have been found in asthma from bronchial biopsies (Macfarlane,
Kon et al. 2000) and in post-mortem tissue (Koshino, Teshima et al. 1993; Kepley, McFeeley et
al. 2001). In allergic rhinitis they are the main source of histamine during the late phase
response after allergen challenge (Bascom, Wachs et al. 1988).
Timothy SC Hinks 1. Introduction
7
Neutrophils
Neutrophils are the first cell type to be recruited to the airways during allergen challenge and
have been implicated in the pathology of nocturnal asthma, and sudden asthma death (Sur,
Crotty et al. 1993). Neutrophilic asthma defined as >61% neutrophils in induced sputum (Belda,
Leigh et al. 2000) affects between 20 and 30% of adults with persistent asthma (Green,
Brightling et al. 2002; Simpson, Scott et al. 2006), being more common in older people, more
severe asthma and those with poor response to corticosteroids (Simpson, Phipps et al. 2009).
Neutrophil survival is prolonged within the airways by antiapoptotic factors which are currently
unidentified, but known to be quantitatively different in more severe asthma (Uddin, Nong et al.
2010). As neutrophilic inflammation characterises bronchiectasis and is correlated with bacterial
load it is likely that bacterial colonisation may be a precursor to persistent airway neutrophilia
(Angrill, Agusti et al. 2001). In turn neutrophil products impair mucociliary clearance through
induction of mucus hypersecretion (O'Donnell, Breen et al. 2006) and a reduction in ciliary
function (Amitani, Wilson et al. 1991), leading to a vicious cycle of airways inflammation
(Simpson, Phipps et al. 2009). Neutrophils are recruited to the airways by such chemotactic
mediators as IL-8 (CXCL8) and CXCL1 (GRO-alpha) and release mediators such as neutrophil
elastase, MMP-9 and oxidative free radicals which can be directly destructive to airway tissue
and are likely to contribute to the development of irreversible airflow obstruction. These products
are elevated in the airways of neutrophilic asthma, and can recruit and activate further
neutrophils in a self-maintaining cycle (Simpson, Phipps et al. 2009). Ironically the mainstay
treatments in asthma pharmacotherapy almost certainly contribute to airway neutrophilia, as
both corticosteroids (Saffar, Ashdown et al. 2011) and β2-agonists prolong neutrophil survival
(Perttunen, Moilanen et al. 2008).
Macrophages
Macrophages are the predominant immune cells in the airways. Macrophages which secrete
type 2 cytokines (IL-4 and -13) and chemokines (the CCR4 ligands CCL17 and CCL22) have
been termed alternatively activated (M2) macrophages. Using animal models, M2 macrophages
have been implicated in allergic lung inflammation. We have recently shown that although the
full M2 phenotype is not seen in human lungs, asthma is characterised by an increased
expression of CCL17 in alveolar macrophages and its expression correlates with eosinophilia
(Staples, Hinks et al. 2012). Macrophages express the low affinity IgE receptor which is up-
regulated in asthma, implying some involvement in atopic allergic responses and expression of
eicosanoids, superoxide, platelet activating factor and granulocyte macrophage colony
stimulating factor are all increased in alveolar macrophages in asthma (Arnoux, Duval et al.
1980; Godard, Chaintreuil et al. 1982; Damon, Chavis et al. 1983; Capron, Jouault et al. 1986).
Timothy SC Hinks 1. Introduction
8
Inflammatory mediators
Asthma involves dysregulation of a complex, integrated immune system in which different cell
types contribute to an inflammatory network orchestrated by an array of pleotropic and
redundant inflammatory mediators. These include leukotrienes, prostanoids, nitric oxide,
platelet-activating factor, bradykinin, chemokines and cytokines (Holgate 2011). Leukotrienes
are eicosanoid lipids synthesised from arachidonic acid by 5-lipoxygenase; they are potent
bronchoconstrictors and have been successfully targeted therapeutically by the leukotriene
receptor antagonists zafirlukast and montelukast (Dempsey 2000; Laidlaw and Boyce 2012).
Prostanoids are another class of pro-inflammatory eicosanoid, generated by cyclooxygenase
and include the prostaglandins, thromboxanes and prostacyclins and may play a role in aspirin
sensitive asthma. Nitric oxide acts as a non-adrenergic, non-cholinergic neurotransmitter in the
airways, can mediate vasodilation, and is a useful biomarker of airway eosinophilia (Taylor,
Pijnenburg et al. 2006). Cytokines are peptide mediators released from inflammatory cells,
which are important in signalling between cells (Barnes, Djukanovic et al. 2003; Holgate 2011)
and include the interleukins (ILs) which act to stimulate, regulate, or modulate lymphocytes such
as T cells (Murphy, Travers et al. 2008). Over 50 cytokines have been identified, which may
have pro- or anti-inflammatory roles, or have actions which are context dependent (Murdoch
and Lloyd 2010), and interact in complex networks. Much attention in allergy research has
focused on the TH2 cytokines IL-4 (critical for IgE class switching in B cells(Lebman and
Coffman 1988)) and IL-5 (important for the terminal differentiation, survival and activation of
eosinophils (Sanderson 1992)). Understanding of these inflammatory networks has led to the
recent development of two new therapeutic strategies in asthma, ie monoclonal antibodies to
IgE (omalizumab) and IL-5 (mepolizumab).
Innate responses
Polymorphisms in toll-like receptors (TLR)s and associated molecules suggest that in addition to
the clear role of adaptive immunity, differences also in innate immunity may contribute to
asthma pathogenesis (Lazarus, Raby et al. 2004; Moller-Larsen, Nyegaard et al. 2008;
Bjornvold, Munthe-Kaas et al. 2009; Bjornsdottir, Holgate et al. 2011). Gene expression profiling
of peripheral blood mononuclear cells (PBMC) during asthma exacerbations showed activation
of innate pathways including TLR1, 2, 3 and type I IFN (Bjornsdottir, Holgate et al. 2011).
DerP2, a major component of house dust mite allergen (HDM), shares structural homology with
the lipopolysaccharide (LPS) binding component of TLR4, giving it intrinsic adjuvant properties
which may explain the high frequency of HDM sensitisation (Trompette, Divanovic et al. 2009),
whilst cockroach frass contains a TLR2 agonist which can directly activate neutrophils(Page,
Lierl et al. 2008). Indeed sputum from neutrophilic asthmatics has higher expression of various
molecules of innate immunity including TLR2, 4 and IL-8(Simpson, Grissell et al. 2007).
Timothy SC Hinks 1. Introduction
9
Airways remodelling in asthma
Inflammatory cells do not function in isolation, but interact continually with structural tissues of
the airways. Asthma typically involves characteristic changes to the bronchial epithelium
including epithelial metaplasia, thickening of the subepithelial basal lamina, increased number
of myofibroblasts and other evidence of airway remodeling such as hypertrophy and hyperplasia
of airway smooth muscle, mucous gland hyperplasia, angiogenesis and an altered extracellular
matrix (Holgate 2008). These features, along with upregulation of epidermal growth factor
receptors and reduced makers of cell proliferation, suggest that the asthmatic airway epithelium
is chronically injured. Causative factors in this injury include inhaled allergens, viral infections or
airway pollution. In response to chronic injury the epithelium can secrete growth factors such as
transforming growth factor-β (TGFβ), platelet-derived growth factor, and fibroblast growth
factors, which act on surrounding stromal cells to induce the features of airway remodelling such
as goblet cell hyperplasia, smooth muscle hypertrophy and myofibroblast differentiation. The
interaction between such a susceptible epithelium and TH2-mediated inflammation, altering
communication between the epithelium and the underlying mesenchyme, has led to the concept
of the ‘epithelial mesenchymal trophic unit’ in which these interplays lead to disease
persistence, airway remodelling and refractoriness to corticosteroids (Holgate, Lackie et al.
2001). This interplay between mesenchyme and epithelium has been underlined by the recent
discovery that smooth muscle contraction alone, induced by methacholine challenge, is
sufficient to induce an increase in subepithelial collagen-band thickness, a marker of airway
remodelling (Grainge, Lau et al. 2011).
In summary underlying the spectrum of disorders classified as asthma are a wide range of
distinct pathological changes, arising from the complex interplay of several intricate biological
systems, including chronic injury and activation of structural cells, innate cells and the cells of
the adaptive immune system. The preeminent effector and regulatory cells of the cellular
adaptive immune system are T lymphocytes.
T lymphocytes (T cells)
T cells are defined by their surface expression of clonally distributed T cell receptors (TCRs)
and play a central role in cell mediated immunity. They develop from progenitors that are
derived from the pluripotent haematopoietic stem cells in the bone marrow and migrate through
the blood to the thymus, where they mature, and it is for this reason that they are called thymus-
dependent (T) lymphocytes or T cells (Murphy, Travers et al. 2008). T cells comprise a
heterogeneous spectrum of subsets with differing expression of TCR classes – TCRαβ or
TCR -, CD4 and CD8 lineage markers and other surface phenotypes, and very distinct
immunological functions. A fundamental dichotomy amongst the major class of TCRαβ+ T cells
is determined by expression of either the CD8 co-receptor, enabling these cytotoxic T cells
directly to kill cells with intracellular infections, or the CD4 co-receptor defining the T helper (TH)
cell subset which provides essential additional signals to activate B cells or macrophages to
Timothy SC Hinks 1. Introduction
10
stimulate antibody production or increased cell killing respectively (Murphy, Travers et al. 2008),
although there are important exceptions to this general scheme (Mumberg, Monach et al. 1999).
Amongst TCRαβ+ CD4+ T cells different cells have differing cytokine secretion profiles which
have been used to define TH1, which activate infected macrophages and provide co-stimulation,
and TH2 cells, which primarily activate naïve B cells to produce antibody (Mosmann, Cherwinski
et al. 1986). In recent years many additional T cell subsets have been described including
immunoregulatory regulatory T cells (Treg) (Thornton and Shevach 1998), TH17 cells (Park, Li
et al. 2005), innate-like lymphocytes such as iNKT cells (Taniguchi, Koseki et al. 1996), and
MAIT cells (Tilloy, Treiner et al. 1999).
The importance of T cells in asthma
T cells are widely recognised as orchestrators of the immune response in asthma. They are
increased in asthmatic airways in correlation with activation status (Azzawi, Bradley et al. 1990;
Walker, Kaegi et al. 1991; Bentley, Menz et al. 1992; Larche, Robinson et al. 2003). Analysis of
sibling pairs revealed genetic linkage between specific IgE responses and a gene in the TCR-α
gene complex on chromosome 7 (Moffatt, Hill et al. 1994). Furthermore, T cells can influence
the function of many inflammatory cells including mast cells and eosinophils through the
production of a group of pro-inflammatory cytokines in the IL-4 gene cluster on chromosome
5q31–33, which tend to exacerbate allergic responses (Holgate 1999). These cytokines define a
distinct T cell subset, T helper-2 (TH2) cells, which in the early 1990s were shown by Robinson
et al to predominate amongst allergic asthmatics (Robinson, Hamid et al. 1992). The TH2
cytokines all play key roles in allergic asthma: IL-4 is important for allergic sensitization and IgE
production, and IL-5 is crucial for eosinophil survival, whilst IL-13 has pleiotropic effects in the
lungs including a central role in the development of airway hyper-responsiveness and tissue
remodelling (Holgate 2008; Lloyd and Hessel 2010). Allergen challenge in asthmatics can
induce airway recruitment of activated TH2 cells, with concomitant increase in TH2 cytokines and
eosinophilia(Larche, Robinson et al. 2003). Conversely, interferon (IFN)- secreting Th1 cells
which antagonise Th2 mediated responses are generally thought not to play a major role in
allergic airways inflammation(Holgate 1999).
Beside theoretical considerations and observational associations, what other evidence is there
of a causal role for aberrant T cells responses in the pathogenesis of asthma? Till et al
performed segmental bronchoscopic allergen challenge of house dust mite-sensitive asthmatics
and healthy controls. Allergen challenge increased BAL and peripheral T cell proliferation and
IL-5 production in asthmatics and these BAL responses correlated with the degree of BAL
eosinophilia, implying that allergens induce pathogenic allergen-specific TH2 responses in the
airways(Till, Durham et al. 1998).
Evidence that T cells may be sufficient to provide a trigger for the development of asthma
comes from reports of asthma resulting from the adoptive transfer of T cells in autologous bone
Timothy SC Hinks 1. Introduction
11
marrow transplant (BMT) recipients. Rietz et al report two individuals who developed asthma
after BMT from a human leukocyte antigen (HLA) identical sibling with asthma, including the
acquisition of measureable airflow obstruction, bronchial tissue eosinophilia, and clinical
response to inhaled steroids and bronchodilators (Rietz, Plummer et al. 2002). Hallstrand et al
subsequently followed 5 long-term survivors of BMT received from allergic donors finding high
IgE levels frequently persisted and a high rate of new sensitisation occurred, leading to rhinitis
and – in 4/5 individuals – asthma (Hallstrand, Sprenger et al. 2004).
Whilst transfer of this phenotype might be caused by transfer of T cells, B cells or hematopoietic
stem cells, conversely an interventional trial of anti-CD4 monoclonal antibody (Keliximab)
increased lung function in asthmatics and showed slight, albeit non-significant, trends towards
improved symptoms, providing intriguing evidence in humans that T cells are at least one
necessary component for the development of asthma(Kon, Sihra et al. 1998).
Interleukin-17
Interleukin-17 (IL-17, also called IL-17A) is a cytokine produced by activated memory T cells,
and other tissues (Fossiez, Djossou et al. 1996). It was first identified in 1993 by cloning the
human homolog of murine cytotoxic T lymphocyte associated antigen (mCTLA8), and was
found to be produced on activation of T cells by phorbol myristate acetate and
ionomycin(Fossiez, Djossou et al. 1996). It is a disulfide-linked homodimeric glycoprotein of 155
amino acids(Yao, Fanslow et al. 1995), acting as a 35kDa homodimer (Kolls and Linden 2004),
and encoded at gene locus 6p12 (Moseley, Haudenschild et al. 2003). IL-17A is now
recognised to be the prototypic member of the IL-17 cytokine family which contains 5 further
cytokines which were identified by gene database searches, cloned and named IL-17B to IL-
17F(Li, Chen et al. 2000; Fort, Cheung et al. 2001; Lee, Ho et al. 2001; Starnes, Robertson et
al. 2001; Hurst, Muchamuel et al. 2002). Whilst all six have some degree of structural homology
– a common cysteine knot formation (Hymowitz, Filvaroff et al. 2001) – they are otherwise a
genetically and functionally divergent group. IL-17A and IL-17F are most related, both being
encoded on the same chromosome. As they are located only 45 kBP apart, they are probably
co-regulated, indeed both being induced by IL-23(Kolls and Linden 2004). They also share the
closest (40-55%) sequence homology (Kolls and Linden 2004) and have functional similarities,
as they both induce a neutrophil response (Kolls and Linden 2004).
In contrast to the similar structures and functions of IL-17A and IL-17F, the other family member
(IL-17B to E) are more diverse. They have lower (17-29%) sequence homology to IL-17A, and
are encoded on four different chromosomes(Kolls and Linden 2004). Neither IL-17B or IL-17C
are expressed in the lung (Li, Chen et al. 2000) and IL-17D is expressed on the endothelium
rather than epithelium. Whilst IL-17E is expressed in lung tissue, its sequence is the most
distantly related to IL-17A, and this cytokine is now better known as IL-25. IL-25 is considered a
TH2-type cytokine (Kolls and Linden 2004) which has been shown to suppress IL-17A
Timothy SC Hinks 1. Introduction
12
responses (Bettelli, Korn et al. 2008) and induce eosinophilic inflammation(Letuve, Lajoie-
Kadoch et al. 2006).
Receptors for IL-17 so far identified include IL-17R, IL-17RH1, IL-17RL (receptor like), IL-17RD
and IL-17RE, all of which are type-1 transmembrane receptors whose differences result from
alternative splicing (Kolls and Linden 2004). They are ubiquitously expressed on a variety of
tissues including lung, but also cartilage, bone, meniscus, brain, hematopoietic tissue, kidney,
skin and intestine; and on a variety of cell types including epithelial cells, fibroblasts, B and T
cells, myelomonocytic cells and marrow stromal cells (Moseley, Haudenschild et al. 2003).
Receptor engagement on stromal cells leads, via the adaptor ACT1/CIKS and TRAF-6, to
activation of the transcription factor NF-kB and the Jnk kinases (Schwandner, Yamaguchi et al.
2000), inducing secretion of pro-inflammatory cytokines including IL-6, IL-8 (CXCL8), CXCL2,
PGE2 and G-CSF (Fossiez, Djossou et al. 1996) which are chemotactic for neutrophils
(Sergejeva, Ivanov et al. 2005; Fujiwara, Hirose et al. 2007; McKinley, Alcorn et al. 2008), as
well as the cytokine IL-22 which in turn induces the antimicrobial peptide human β-defensin 2
(Wiehler and Proud 2007).
The T helper 17 subset
Expression of IL-17, in the absence of IFN-, defines a recently described subset of CD4+ T
helper lymphocytes called T helper-17 (TH17) lymphocytes (Park, Li et al. 2005). They comprise
a distinct T cell lineage, that is not dependent on Th1 and Th2 associated transcription factors T
Bet or GATA3 (Park, Li et al. 2005) but on expression of the nuclear transcription factor retinoic
acid-related orphan nuclear hormone receptor (ROR)C (or its homolog RORt in mice), which
induce IL-17A and IL-17F (Ivanov, McKenzie et al. 2006). Importantly the predominantly pro-
inflammatory TH17 cells share a reciprocal developmental pathway with FOXP3+ regulatory T
cells (Treg) implying this dichotomy may have evolved to induce or regulate tissue inflammation
(Bettelli, Carrier et al. 2006).
In humans TH17 can be induced in vitro by culture of naïve T cells with IL-21 and TGFβ, or from
central memory T cells by IL-1β and TGFβ (Yang, Anderson et al. 2008). Maintenance of the
TH17 cell population may depend on the presence of IL-23 (Yang, Anderson et al. 2008).
Emerging animal data suggest reciprocal developmental relationships between TH17 and Treg
(Mucida, Park et al. 2007; Lochner, Peduto et al. 2008) with antagonistic functions of the
RORT and Forkhead box P3 (FOXP3) transcription factors. In the gastrointestinal mucosa
TH17 can induce chronic inflammation(Leppkes, Becker et al. 2008) and it has been suggested
that regulation may be influenced by the mucosal microflora (Ivanov, Frutos Rde et al. 2008;
Zhou, Lopes et al. 2008).
Timothy SC Hinks 1. Introduction
13
Interleukin-17, TH17 cells and asthma
TH17 cell have been linked to neutrophilic pulmonary inflammation in both human asthma
(Molet, Hamid et al. 2001) and mouse models of allergic inflammation (Park, Li et al. 2005). In
the airways IL-17 primarily acts on stromal cells to induce cytokines and chemokines including
IL-6, -8 (CXCL8), CCL26 (eotaxin-3), CXCL1,and CXCL2 (Fossiez, Djossou et al. 1996; Wang,
Voo et al. 2010), which are chemotactic for neutrophils (Sergejeva, Ivanov et al. 2005; Fujiwara,
Hirose et al. 2007; McKinley, Alcorn et al. 2008) (See Figure 1.1). Mouse models (Schnyder-
Candrian, Togbe et al. 2006; Fujiwara, Hirose et al. 2007; McKinley, Alcorn et al. 2008), human
genetic associations (Hizawa, Kawaguchi et al. 2006; Kawaguchi, Takahashi et al. 2006; Chen,
Deng et al. 2010; Lluis, Schedel et al. 2011) and studies of protein and messenger RNA
(mRNA) expression in sputum or bronchoalveolar-lavage (BAL) (Molet, Hamid et al. 2001;
Chakir, Shannon et al. 2003; Bullens, Truyen et al. 2006), have implicated IL-17 in the
pathogenesis of asthma and bronchial hyper-reactivity.
Timothy SC Hinks 1. Introduction
14
Figure 1.1 IL-17 and TH17 cells in the immune response
A schematic diagram of the induction of TH17 cells in an immune response to infection in
humans. Infection leads to recognition by the innate immune system: macrophages and
complement leading to endothelial cell activation which favours adhesion and extravasation of
other immune cells. There is also early recognition and activation of innate-like lymphocytes
including invariant iNKT and MAIT leading to early production of IL-17. The adaptive immune
system is also triggered, first by the activation of dendritic cells which migrate to draining lymph
nodes where they present antigen to naïve T cells. Primed T cells mature into effector T cells
(Teff), whose fate is determined by the cytokine milieu. In the presence of TGFβ alone the
express the nuclear transcription factor FOXP3, becoming immunoregulatory inducible T reg. By
contrast in the presence of TGFβ and IL-21 the transcription factor RORC2 is expressed. This
physically interacts with, and displaces FOXP3 from nuclear binding sites, and induces an IL-
17+ IFNγ- TH17 phenotype. TH17 cells secrete IL-21 which promotes expansion of the TH17 cell
pool in an autocrine manner. The TH17 population may also be expanded by recruitment of
memory T cells (Tmem) under the influence of IL-1β and IL-6. TH17 cell frequencies are
maintained by IL-23. TH17 cells secrete IL-17(A) and IL-17F which act on stromal cells such as
fibroblasts, epithelia and keratinocytes to induce secretion of numerous pro-inflammatory
chemokines, including IL-6, IL-8 (CXCL8) and CXCL2 which recruit neutrophils (Nϕ), and G-
CSF, GM-CSF which up-regulate production of macrophages (Mϕ). Induction of IL-22 induces
further inflammation by induction of acute phase proteins, epidermal hyperplasia, and the
antimicrobial peptide β-defensin, which may be considered an end effector molecule TH17 cells.
Timothy SC Hinks 1. Introduction
15
Molet reported an increased number of IL-17+ cells in sputum and BAL from six asthmatics
(Molet, Hamid et al. 2001), and later the same group reported an increase in IL17+ staining in
the submucosa and epithelium of nine moderate-severe asthmatics (Chakir, Shannon et al.
2003), whilst others reported increased numbers of submucosal IL-17 in mild-moderate but not
severe asthma (Doe, Bafadhel et al. 2010). Furthermore there have been reports of a
correlation between whole sputum IL-17 mRNA and bronchial hyper-responsiveness (Barczyk,
Pierzchala et al. 2003), or the presence of asthma (Zhou, Sun et al. 2005; Bullens, Truyen et al.
2006), and also of increased IL-17 mRNA in bronchial biopsies (Vazquez-Tello, Semlali et al.
2010; Howarth 2012).
IL-17 and TH17 cells in murine models of allergic airways disease
Despite this extensive body of literature regarding IL-17 and TH17 cells in animal models, very
little comparable human data have been obtained to date, and it is currently unknown whether
TH17 cells are involved in human asthma. Indeed, even if IL-17 levels are found to be elevated
in asthma, it could have a wide variety of potential cellular sources; thus, not only TCRαβ+ T-
cells, but also TCR+ T-cells(Lochner, Peduto et al. 2008), eosinophils (Molet, Hamid et al.
2001) and macrophages (Song, Luo et al. 2008), and even B cells (Vazquez-Tello, Halwani et
al. 2012) can potentially secrete IL-17. Moreover it is unknown which putative surface markers
(Acosta-Rodriguez, Rivino et al. 2007; Cosmi, De Palma et al. 2008; Pene, Chevalier et al.
2008) identify pulmonary TH17 cells or whether TH17 frequencies and functions are associated
with distinct asthma phenotypes, such as the neutrophilic forms where it is frequently
hypothesised to be significant.
Regulatory T cells
The differentiation of TH17 cells is closely related to that of the functionally antagonistic
regulatory T (Treg) cell subset (Bettelli, Carrier et al. 2006). Tregs have been identified in mice
and humans which are believed to be essential for regulating adaptive immune responses,
regulating the host response to infection, maintaining self-tolerance and preventing autoimmune
diseases (Takahashi, Kuniyasu et al. 1998; Belkaid, Piccirillo et al. 2002). Naturally occurring,
thymic derived CD4+CD25+ Treg cells (syn: natural Tregs) inhibit effector functions of other
immunocytes, eg CD4+ and CD8+ T cells (Sakaguchi, Sakaguchi et al. 1995; Thornton and
Shevach 1998; Baecher-Allan, Brown et al. 2001; Murakami, Sakamoto et al. 2002). Arising
from the thymus, they enter peripheral tissues where they suppress the activation of other self-
antigen–reactive T cells (Bluestone and Abbas 2003). In murine models, they suppress T-cell
responses to several intracellular pathogens, and their depletion in vivo leads to increased
immune-mediated tissue pathology. Natural Tregs require antigen-specific T cell receptor (TCR)
mediated activation, but effector function is non-specific (Liu, Putnam et al. 2006).
Timothy SC Hinks 1. Introduction
16
Different chemokine receptors determine Treg homing to distinct tissues such as lymphoid or
non-lymphoid tissues, or sites of inflammation. Treg act via a number of mechanisms including
secretion of IL-10, TGFβ, IL-35, inhibition of dendritic cell (DC) maturation via surface CTLA4,
direct granzyme and perforin mediated killing of mature DCs, inhibition of priming effector CD4+
T cells or tumour specific CD8+ T cells, and metabolic inhibition of effector T cells via adenosine
and cyclic adenosine monophosphate (cAMP) (Campbell and Koch 2011). Differentiation of
Treg can be modulated by cytokines, steroids, sphingolipids and vitamin A and D metabolites.
IL-2 plays a particularly important role, signalling in a paracrine fashion via the IL-2 receptor
(CD25) to promote Treg survival and proliferation. Treg development can be inhibited by IL-4
with TGFβ, or by IFN-α, and IFN-β. IL-6 inhibits FOXP3 expression and induces TH17 cells,
whilst TNF can potentiate Treg function(Campbell and Koch 2011).
In human disease dysregulation of FOXP3 Treg has been implicated in autoimmunity,
lymphoproliferative disease, type I diabetes mellitus, systemic lupus erythematosus, rheumatoid
arthritis and multiple sclerosis (Campbell and Koch 2011). Treg have also been studied in a
number of chronic infectious diseases such as tuberculosis(Guyot-Revol, Innes et al. 2006),
leishmaniasis (Belkaid and Rouse 2005), bacteria, viruses, parasites and fungi (Mills 2004)
where they may play a role both in limiting immunopathology, but also in maintaining microbial
persistence.
Regulatory T cells in asthma
FOXP3+ Treg have been shown to be present in the bronchial mucosa in infants and were
primarily located within isolated lymphoid follicles of bronchus-associated lymphoid tissue(Heier,
Malmstrom et al. 2008). Lin et al found that asthmatic children had lower FOXP3 levels in
peripheral blood(Lin, Shieh et al. 2008), whilst in a longitudinal study of 18 severe asthmatics
with frequent exacerbations, peripheral blood and sputum Treg were decreased in frequency
and function (blood) during exacerbations in severe asthma (Mamessier, Nieves et al. 2008).
A key effector mechanism for Treg is production of the anti-inflammatory cytokine IL-10. There
is evidence that treatment-refractory asthmatics have impaired steroid-induced IL-10 production
(Holgate and Polosa 2006), and that blood levels of IL-10 correlate inversely with disease
severity in atopic asthma(Matsumoto, Inoue et al. 2004; Hawrylowicz 2005; Matsumoto, Inoue
et al. 2008). Furthermore in atopic individuals IL-10 secreting CD25+ Treg can be induced by
immunotherapy (Ling, Smith et al. 2004) whilst Treg can suppress allergen-activated IL-4 cells,
again via IL-10 and transforming growth factor-β (TGFβ)(Robinson, Larche et al. 2004).
In mouse models of Treg depletion or adoptive transfer Treg numbers correlate negatively with
bronchial hyper-reactivity (Hawrylowicz 2005; Kearley, Barker et al. 2005; Lewkowich, Herman
et al. 2005; Kearley, Robinson et al. 2008). Treg have also been induced by heat killed
Mycobacterium vaccae and inhibited AHR via the induction of TGFβ and IL-10 (Zuany-Amorim,
Timothy SC Hinks 1. Introduction
17
Sawicka et al. 2002; Robinson, Larche et al. 2004). Indeed a trial of M. vaccae in humans
suggested a trend towards a decrease in the late asthmatic response, although this did not
reach significance (Camporota, Corkhill et al. 2003). Likewise murine Treg can also be induced
by dendritic cells transfected with DerP2 DNA. These Treg can suppress TH2 responses,
allergen specific CD4+ T cell responses, and AHR (Wu, Bi et al. 2008). In a rat model of chronic
aeroallergen exposure Treg are induced in the airway mucosa and inhibit subsequent T cell
activation (Strickland, Stumbles et al. 2006).
There have recently been further reports of Treg in peripheral blood in relation to asthma or
allergy. McCloughlin et al studied Treg frequencies in infants prospectively from birth. Treg
frequencies increased over the first two years of life, although their functional capabilities did not
change. Treg frequencies at birth did not predict development of subsequent allergy, but by one
and two years of age Treg frequencies and suppressive function were associated with reduced
allergic sensitization, which appeared to be mediated by IL-10 (McLoughlin, Calatroni et al.
2012). Wang found lower Treg frequencies in peripheral blood in 20 asthmatics (Wang, Lin et al.
2009), and although Provoost et al found no difference in CD4+25HiFOXP3+cell frequencies in
blood between asthma and health, they observed lower FOXP3 expression with these Tregs in
adult asthmatics (Provoost, Maes et al. 2009). Ex vivo steroid treatment of stimulated PBMC
increased the anti-inflammatory ratios of FOXP3/GATA-3, FOXP3/T-bet, and FOXP3/RORC2
(Provoost, Maes et al. 2009).
In summary, a wealth of mouse data suggest a protective role for Treg in asthma, but with the
exception of Mamessier’s excellent work, published human data are predominantly only from
peripheral blood, and a systematic cross sectional analysis of airway Treg, particularly in severe
asthma, is currently lacking.
CD8+ T cells and asthma
In order to place T helper subsets in context I have also undertaken an analysis of cytotoxic
(CD8+) T cells in asthma. In comparison with CD4+ T cells, there has been much less research
on the role of CD8+ T cells in human asthma. Bronchoscopy studies in smokers have reported
increased epithelial CD8+ T cell infiltration in subjects with mild airflow limitation compared with
those with chronic bronchitis alone (Fournier, Lebargy et al. 1989), and an inverse relationship
between CD8+ T cells and FEV1 in chronic obstructive pulmonary disease
(COPD)(O'Shaughnessy, Ansari et al. 1997). A similar increase in CD8+ T cells was found in
surgical lung specimens from subjects with COPD compared with healthy smoker s(Saetta, Di
Stefano et al. 1998).
A post mortem study of seven subjects who died of asthma found higher frequencies of
peribronchial CD8+ T cells, compared with subjects who died of other causes. These cells were
activated, expressing CD25, perforin, IL-4 and IFN-, with a higher IL4/IFN- ratio (O'Sullivan,
Timothy SC Hinks 1. Introduction
18
Cormican et al. 2001). Given the circumstances of their deaths it is likely these cells were
responding to presence of an acute viral infection. Krug et al studied BAL T cells before and
after allergen challenge in 11 subjects with mild atopic asthma. They found no differences in
CD8+ cell frequencies at baseline, but there was a significant fall in the proportion of T cells
secreting IFN-γ and IL2 in the asthmatics after allergen challenge, in both CD4+ and CD8+
subsets(Krug, Erpenbeck et al. 2001). Van Rensen et al observed a correlation (r=-0.39)
between CD8+ T cell frequencies in baseline bronchial biopsies from 32 asthmatics and their
subsequent rate of decline in post-bronchodilator FEV1 (van Rensen, Sont et al. 2005). In 21
infants with wheeze (median age 15.4 months) Arnoux found increased numbers of CD8+ cells
in BAL compared with non-wheezing controls (Arnoux, Bousquet et al. 2001). It is not clear
whether a viral infection might have induced these responses as all children had a history of an
acute viral exacerbation, but the bronchoscopies were performed during a period of clinical
stability, and no respiratory viruses were detectable at the time using immunofluorescence.
As with the TH1/TH2 dichotomy amongst CD4+ T cells, CD8+ T cells form functionally similar
subsets with similar cytokine profiles known as Tc1 and Tc2 (Mosmann, Li et al. 1997). Cho et
al found increased frequencies of both CD4+ and CD8+ sputum T cells spontaneously secreting
IL-4, -5 and IFN- in nine subjects with mild-moderate atopic asthma (Cho, Stanciu et al. 2005).
This increase was related to disease severity, and this association was stronger for CD8+ than
CD4+ cells. Also in peripheral blood Magnan et al found an increase in IFN--CD8 cells which
was related to asthma severity, to bronchial hyper-responsiveness, to blood eosinophilia and to
peripheral blood IL-12 (Magnan, Mely et al. 2000).
In summary little is known about CD8+ T cells in asthma, but what data there are tend to imply a
pathological role for CD8+ cells (Betts and Kemeny 2009).
Mucosal Associated Invariant T (MAIT) cells
T cell immunology is a rapidly evolving field with many new T cell subsets identified in recent
years (Bluestone and Abbas 2003; Shevach 2006; Schmidt-Weber, Akdis et al. 2007). Perhaps
the most exciting of these has been the recognition of distinct classes of innate-like lymphocytes
(Arase, Arase et al. 1993) whose role in airways disease has been a subject of some
controversy. I have taken the opportunity provided by my focused study on TH17 cells to
undertake the first analysis in the airways of a recently discovered class of innate-like
lymphocytes: Mucosal associated invariant T cells (MAIT). This section reviews the current
literature on these cells.
Innate-like lymphocytes
Most T cells have diverse T cell receptors (TCRs) due to stochastic recombinations of V, D and
J segments, with additional random trimming or addition of nucleotides at the junctions(Tilloy,
Treiner et al. 1999). Some “invariant” lymphocyte subsets have more restricted TCRs, namely
Timothy SC Hinks 1. Introduction
19
B1 B cells, some T cells, and the CD1d restricted invariant natural killer T (iNKT) cells (Tilloy,
Treiner et al. 1999). Invariant lymphocytes migrate rapidly to the site of acute inflammation,
have a memory phenotype in the absence of deliberate immunization, and respond rapidly to
challenge, eg with secretion of massive amounts of cytokines. They are therefore also often
called ‘innate lymphocytes’ (Treiner, Duban et al. 2005). Innate T cells in humans and mice
comprise CD1d restricted iNKT and MR1 (MHC related-1) restricted MAIT cells, both of which
express “invariant” TCRs that are conserved between species (Treiner and Lantz 2006) and
recognise nonpolymorphic antigen presenting molecules (ie CD1d and MR1, respectively)
(Porcelli, Yockey et al. 1993). Both iNKT and MAIT cells are believed to react to
phylogenetically-conserved antigens, and both subsets are thought to play key regulatory roles
in immunity (Treiner and Lantz 2006).
Mucosal associated invariant T cells
The canonical TCR for MAIT cells was first identified in 1993 by Porcelli et al who noted that
many people expressed an identically rearranged TCRα chain: Vα7.2-Jα33 (Porcelli, Yockey et
al. 1993). In 1999 Olivier Lantz et al were the first to describe this Vα7.2-Jα33 segment as
defining a new subset of T cells, found in humans, mice (which express the homologue Vα19-
Jα33), and cattle, with a complementarity determining region (CDR)3 of constant length. MAIT
comprise up to 15% of human peripheral blood DN cells (0.1-0.2% of all T cells). They were
initially found to predominantly have a double negative (DN, frequency 1/10) or CD8αβ
phenotype (frequency 1/50). They have an activated/memory phenotype: CD45RAloCD45RO+
(Tilloy, Triener et al. 1999). They are also CD27+ and CD28+, NKR-P1A+, α4β7+ CD56– CD57–
(Treiner, Duban et al. 2005) CD95HiCD62LLo (Dusseaux, Martin et al.). Expression of
α4β7integrin (Treiner, Duban et al. 2005) and the chemokine receptor expression pattern
CCR9IntCCR7-CCR5HiCXCR6HiCCR6Hi (Dusseaux, Martin et al.) enables them to home to the
intestine, where they are abundant in lamina propria but virtually absent from epithelium(Treiner,
Duban et al. 2005).
MAIT cells have an oligoclonal Vβ repertoire, as the TCRβ chain preferentially uses human
Vβ13 and Vβ2 segments, which suggests peripheral expansions (Tilloy, Di Santo et al. 1999).
The Vα7.2-Jα33 chain is the product of a single combination event with a CDR3α of defined
length and reading frame. That there is some variability in this junction suggests the
overrepresentation of this rearrangement is the consequence of selection at the protein level
rather than a genetically programmed recombination process (Treiner, Duban et al. 2005).
Greenaway’s recent in silico analysis of invariant TCRs has suggested the mechanisms by
which these limited TCRs can be produced through ‘convergent recombination’ (Greenaway, Ng
et al. 2012). The canonical TCRα amino acid sequences in both iNKT and MAIT cells are
encoded by at least one germline-derived nucleotide sequence in all reported species, thus they
are not due to random recombinations of V and J segments. Furthermore these sequences use
Timothy SC Hinks 1. Introduction
20
an overlap between the Vα and Jα genes, in that some of the nucleotides in key CDR codons
can come from either the V or J genes, and in some cases palindromic additions are possible,
ensuring they are produced by a greater variety of recombination mechanisms.
CD161
MAIT cells have a high surface expression of CD161 (Martin, Treiner et al. 2009; Le Bourhis,
Martin et al. 2010; Dusseaux, Martin et al. 2011). CD161 (KLRB1, NKRP1A) is a C-type lectin
which is part of the NK complex. Amongst CD8+ cells, high CD161 expression is associated
with TH17 differentiation: expression of IL-17, IL-22, RORT and IL-23R. CD161 is also
expressed by most NK cells as well as T cells, most NKT cells and many tissue-infiltrating T
cells (Billerbeck, Kang et al. ; Dusseaux, Martin et al. 2011). For the purposes of this work I
have defined MAIT cells using a combination of expression of TCR-Vα7.2 and CD161,
consistent with these studies.
MAIT cell restriction
MAIT cell selection and expansion was shown to be dependent on β2-microglobulin (β2M) but
not major histocompatibility complex (MHC) II or MHC I, suggesting early on that they were
restricted by a non-classical MHC class 1b molecule (Tilloy, Treiner et al. 1999). This class 1b
molecule was later found to be ‘MHC-related protein 1’ (MR1), a highly conserved monomorphic
MHCI related molecule. MR1 is encoded on chromosome. With a remarkable 90% sequence
identity between mouse and human it is the most highly conserved MCH1 related molecule in
mammals (Brossay, Chioda et al. 1998).
MR1 has four isoforms in humans of which only MR1-A is translated and expressed as a
heterodimer with β2M. Very stringent conservation of the MR1 amino acid sequence, even
distally in the molecule, implies strong evolutionary pressure and the possibility that MR1 is part
of a multi-molecular complex or binds to other receptors and co-receptors(Treiner, Duban et al.
2005). It is has long seemed likely that MR1 has an antigen presenting function (Huang, Gilfillan
et al. 2005). MR1 mRNA expression seems to be ubiquitous, though it is rarely detectable at the
cell surface suggesting it is only surface expressed in the presence of its ligand (Treiner, Duban
et al. 2005). This idea has been supported by murine data in which use of a monoclonal
antibody to stabilise endogenous MR1 at the cell surface increased MAIT cell activation (Chua,
Kim et al. 2011).
MAIT cell ligands
Until recently the ligand for MAIT cells has remained obscure. The presentation pathway of MR1
to MAIT cells is highly evolutionarily conserved (Huang, Martin et al. 2009). MR1 traffics through
endocytic compartments, thereby allowing MAIT cells to sample both endocytosed and
endogenous antigens(Huang, Gilfillan et al. 2008). Using conformation-dependent monoclonal
antibodies to detect surface MR1 Abos et al showed MR1 expression was increased at 26°C,
Timothy SC Hinks 1. Introduction
21
was lost with acid, and independent of the proteasome, suggesting that MR1 binds proteasome-
independent ligands (Abos, Gomez Del Moral et al. 2011). Site directed mutagenesis and
analysis of the MR1 crystal structure suggested that only two residues, on either side of the
MR1 cleft, are essential for TCR activation. This, and the relatively rigid TCR, is characteristic
of innate receptors evolved to recognize a very limited range of antigens (Reantragoon, Kjer-
Nielsen et al. 2012).
Kjer-Nielsen et al have recently shown that B2 vitamin derivatives can occupy, though only
partially, the MAIT TCR binding grove. These authors were able to obtain a crystal structure of
6-formyl pterin, a folic acid (vitamin B9) metabolite, bound to MR1, showing the pterin ring
sequestered within MR1 (Kjer-Nielsen, Patel et al. 2012). It seems unlikely this is the natural
ligand for MAIT as the binding was irreversible, left much of the binding grove unoccupied, and
the complex did not activate MAIT cells. However this group have also shown binding of related,
bacterially-derived vitamin B derivatives, such as those originating from the bacterial riboflavin
(vitamin B2) biosynthetic pathway, which can activate MAIT cells. As many microbes have
unique synthetic pathways for vitamins, it seems likely that MAIT cells may recognise
microbially-derived products of vitamin biosynthesis as a means of detecting infection.
MAIT cell development
Selection and expansion of MAIT cells depends on B cells, and also on the presence of
commensal flora, as MAIT cells are not present in germ-free mice (Sano, Haneda et al. 1999).
MAIT cell development is a stepwise process, with an intra-thymic selection followed by
peripheral expansion. While MAIT cell development is thymus dependent (absent in nude mice),
they are rapidly exported from thymus as they are not readily detectable in thymus by PCR.
After birth, MAIT cells acquire a memory phenotype and expand dramatically to 1%-4% of blood
T cells (Marks, Ng et al. 2003; Martin, Treiner et al. 2009; Gold, Eid et al. 2012). MAIT cell
frequencies are 5 to 10 fold lower in mice than humans, which is the converse of iNKT cells
(Treiner, Duban et al. 2005).
MAIT Cell function
A high proportion of transgenic MAIT cells express the natural killer receptor NK1.1, and most
have a cell surface phenotype similar to that of Vα14 iNKT cells. They secrete IFN-gamma, IL-4,
IL-5, and IL-10 following TCR ligation. There may be two functionally distinct MAIT cell
populations; NK1.1+ which can’t express IL10 – and are therefore analogous to iNKT cells - and
NK1.1- which express high levels of IL10(Kawachi, Maldonado et al. 2006). MAIT cells also
produce IFN- and Granzyme-B as well as high levels of IL-17 (Dusseaux, Martin et al.).
MAIT cells are believed to play an important role in defence against a range of microbial
infections. They can recognise cells infected with bacteria such as Escherichia coli, Salmonella
typhimurium, and Staphylococcus aureus, and mycobacteria, or yeasts, but not viruses (Dong,
Timothy SC Hinks 1. Introduction
22
Yang et al. 2005; Le Bourhis, Guerri et al. 2011). Gold et al showed that even naïve MAIT cells
from cord blood can recognise Mycobacterium tuberculosis (MTB) infected cells(Gold, Eid et al.
2012), implying they have intrinsic effector function. Gold also showed that MTB-reactive MAIT
cells predominate in uninfected individuals, they respond to MTB-infected MR1-expressing lung
epithelial cells, decrease in PBMC from subjects with active TB, and were enriched in lung
tissue from 2 subjects with pulmonary TB (Gold, Cerri et al. 2010). In vitro MAIT cells can inhibit
growth of Mycobacterium bovis Bacillus Calmette-Guérin (BCG) within macrophages, in a
mechanism dependent on IFN- (Chua, Truscott et al. 2012). Interestingly in this work MAIT cell
responses were not dependent on cognate recognition of MR1 by MAIT cells, but rather on
macrophage secretion of IL-12.
MAIT cells in human disease
MAIT cells are abundant in humans and express tissue homing integrins and chemokine
receptors. They are common in renal and brain tumours and have been found in a number of
inflammatory tissues (Dusseaux, Martin et al. 2011). Their presence in tissues correlates with
pro-inflammatory cytokines (Peterfalvi, Gomori et al. 2008).
MAIT accumulate in some lesions in multiple sclerosis (MS), and also in chronic inflammatory
demyelinating polyneuropathy (Illes, Shimamura et al. 2004). Whilst iNKT cells are reduced in
peripheral blood in MS, Illes et al did not find any decrease in MAIT cells, whilst others have
observed reduced peripheral MAIT frequencies in MS patients in remission, and particularly in
relapse (Miyazaki, Miyake et al. 2011). Conflicting data have been obtained from animal models
of experimental autoimmune encephalomyelitis (EAE) with some finding no evidence of MAIT
involvement (Yokote, Miyake et al. 2008) and others suggesting that MAIT cells inhibit EAE, and
MR1 deficiency increases EAE (Croxford, Miyake et al. 2006).
Some limited data have been obtained from other mouse models of human disease. Data from
collagen induced arthritis (CIA), a mouse arthritis model, suggested a pathogenic role of MAIT
cells, as knock-out of MR1 ameliorated arthritis, whilst reconstitution with MAIT cells induced
severe disease (Chiba, Tajima et al. 2012). Conversely MAIT cells seemed to be protective in a
mouse model of inflammatory bowel disease, as adoptive transfer reduced the severity of the
colitis (Ruijing, Mengjun et al. 2012).
MAIT cells and the lung
To date MAIT cells have not been studied in the human lung, with two exceptions. Sano et al
refer to unpublished data from 2003 of RNA from frozen lung biopsies suggested MAIT cells
might be present in the lung (Sano, Haneda et al. 1999), whilst Dong et al report that MAIT cells
have been observed in lung tissue from 2 subjects with pulmonary TB (Dong, Yang et al. 2005).
No published data are currently available that characterise MAIT cells in relation to human lung
disease.
Timothy SC Hinks 1. Introduction
23
Vitamin D
Vitamin D is a fat soluble vitamin with pleotropic effects on cell differentiation and function.
Vitamin D deficiency has long been known to be associated with increased risk of immune
mediated disease or of impaired cell mediated immunity, such as active MTB infection, whilst
exogenous 1,25-dihydroxy vitamin D(3) (1,25(OH)2D3) can suppress TH1 mediated immune
responses (Ooi, Chen et al. 2012). Vitamin D status has been implicated in asthma
pathogenesis by genetic associations with the vitamin D receptor and by a number of
observational studies, although the results of these have been conflicting. Generally, data
suggest that vitamin D is protective against asthma, but firm conclusions will depend on the
outcome of prospective clinical trials which are currently ongoing (Paul, Brehm et al. 2012).
Nonetheless, several potential mechanisms have been proposed linking vitamin D with asthma,
including direct antiviral properties, enhanced steroid responsiveness and down-regulation of
atopy (Paul, Brehm et al. 2012). Of particular relevance to this thesis, it has been shown
recently that vitamin D has an effect on the number and functions of innate T cells, specifically
iNKT cells. In utero vitamin D deficiency in mice causes a lasting reduction in iNKT cell
frequencies in the progeny, due to increased apoptosis of early iNKT cell precursors in the
thymus(Yu and Cantorna 2011). Genetic deficiency of vitamin D receptor in mice causes a
reduction in iNKT numbers, and impairs the development of experimental airways hyper-
reactivity, which can be rescued by adoptive iNKT cell transfer (Yu, Zhao et al. 2011). Vitamin D
receptor knockout also affected iNKT cell function, as these cells produced less IL-4, -5, -13 and
-17.
These effects have not been investigated in human asthma, and to date there are no published
data on the effect of vitamin D on MAIT cell number or function.
The Microbiome
The innate and adaptive responses within a mucosal immune system are intrinsically related to
the presence of associated microbial flora. In my thesis I have therefore attempted to
characterise these T cell responses in relationship to the airway flora. This sections reviews
current knowledge of the nature of the airway microflora.
The term ‘microbiome’ was coined by Joshua Lederberg, to describe the totality of microbes,
their genomes, and environmental interactions in a particular environment (Highlander 2012).
The emerging use of molecular techniques to identify microbes without the need for traditional
culture techniques has led to a recent, intensive effort to characterise distinct microbial flora and
anatomical niches of the human microbiome (Costello, Lauber et al. 2009; Nelson, Weinstock et
al. 2010). As over 70% of body surface microbes cannot be cultured by standard techniques,
culture is no longer considered the gold-standard method for microbial investigation of the
Timothy SC Hinks 1. Introduction
24
complex microbial populations (Han, Huang et al. 2012). Various molecular techniques have
been developed, including fluorescent in-situ hybridisation (FISH) with flow cytometry or
analysis of terminal restriction fragment length polymorphism (T-RFLP), but these are limited by
difficulties in ascribing definitive taxonomies to highly variable communities, or previously
undiscovered organisms. Instead the new field of metagenomics depends on high throughput
sequencing of entire populations using shot-gun sequencing of whole genomes, a technique
which can detect fungi and viruses as well as bacterial RNA (Han, Huang et al. 2012).
The lung microbiome
The use of culture-based techniques led to the traditional teaching that the human lung is sterile
in health (Laurenzi, Potter et al. 1961; Pecora 1963). This view has been challenged by the use
of culture-independent techniques. Using sequencing of the 16S subunit of ribosomal RNA from
bronchial brushings, Hilty et al found a mean of 2000 bacterial genomes cm-2 in the bronchial
tree; a figure comparable to that in the upper small intestine (Hilty, Burke et al. 2010). This
group also observed an increased abundance of pathogenic proteobacteria, particularly
Haemophilus species in asthma compared with health, with a concomitant decrease in
bacteroidetes and prevotella species. This study was limited by small numbers (only 13 adult
asthmatics), no correlation with clinical or immunological data and the restriction of the
technique to the identification of bacterial species only.
A particular challenge to the analysis of the lung microbiome is posed by the relative
inaccessibility to direct sampling, compounded by the use of highly sensitive DNA amplification
techniques on relatively low biomass samples, leading to a high risk of detecting upper airway
or oral contaminants. Charlson et al compared the different available sampling techniques using
16S RNA sequencing on oral wash, oropharyngeal swabs, nasopharyngeal swabs,
bronchoalveolar lavage and protected bronchial brushing in 6 healthy individuals (Charlson,
Bittinger et al. 2011). Their findings suggested that, in contrast to other organ systems, there is
no unique lung microbiome in health, but rather bacterial communities are indistinguishable from
those of the upper airways, but two to four log lower in biomass. This implies that microbes
present in healthy lungs are likely to be the product of microaspiration, rather than the existence
of independent communities.
The lung microbiome in cystic fibrosis (CF)
In stark contrast to the situation in health, much research in bronchiectasis, particularly that
caused by cystic fibrosis (CF), has shown that complex polymicrobial communities can exist
independently in the lung, maintaining remarkable longitudinal stability despite the use of broad
spectrum antibiotics. It has long been known from culture techniques that patients with CF
acquire infections incrementally over time according to a largely stereotypic sequence, with a
relatively limited set of bacterial species, including Staphylococcus aureus, Haemophilus
influenzae and Pseudomonas aeruginosa(Han, Huang et al. 2012). As disease progresses
Timothy SC Hinks 1. Introduction
25
other opportunistic species such as Burkholderia cepacia complex and Stenotrophomonas
maltophilia are acquired. The use of molecular techniques has revealed a much greater
diversity of species, including anaerobic bacteria such as Prevotella, Veillonella,
Propionibacterium and Streptococcus milleri, as well as a majority of unculturable species, with
over 60 phylogenetically diverse bacterial genera present. These populations remain fairly
unaffected by use of antibiotics (Guss, Roeselers et al. 2011; Daniels, Rogers et al. 2012).
The microbiome in chronic obstructive pulmonary disease (COPD)
In contrast to the microbial diversity observed in CF, the first use of pyrosequencing in BAL and
excised lung tissue in COPD showed very limited community diversity (Erb-Downward,
Thompson et al. 2011). These data were interpreted to imply a core pulmonary bacterial
microbiome including Pseudomonas, Streptococcus, Prevotella, Fusobacterium, Haemophilus,
Veillonella, and Porphyromonas, but as noted this is not consistent with the careful study by
Charlson et al which used many more sampling techniques and methodological controls
(Charlson, Bittinger et al. 2011). Erb-Downard also observed striking micro-anatomic differences
in bacterial communities within different areas of the same lung in subjects with advanced
COPD (Erb-Downward, Thompson et al. 2011). The longitudinal dynamics of these communities
were assessed in a 4 year longitudinal study using molecular typing of sputum from 81 patients
with COPD, which revealed that exacerbations were triggered by acquisition of a new strain of
H. influenzae, M. catarrhalis, or S. pneumoniae, rather than an increase in absolute bacterial
number (Sethi, Evans et al. 2002). It is intriguing to speculate on the source of the microbes, as
16S RNA microarray of cigarettes found 15 different classes of bacteria in cigarettes including
many highly pathogenic organisms like Acinetobacter, Bacillus, Burkholderia, Clostridium,
Klebsiella, Pseudomonas aeruginosa, and Serratia (Sapkota, Berger et al. 2010).
The lung microbiome in asthma
Compared with CF and COPD, much less is known about the microbiome in asthma. The
presence of bacteria might be inferred from the innate immune activation seen in neutrophilic
asthma (Simpson, Grissell et al. 2007). Using T-RFLP Green et al observed H influenza,
Moraxella or Streptococcus in induced sputum from 21/28 severe asthmatics (Green, Kehagia
et al. 2008). In the absence of significant upper airways contamination, cultivable bacteria
represented only 0.1-20% of species, but these three species were the dominant organism in
over half of those colonized. Bacterial colonisation was associated with higher neutrophil count,
longer history of asthma and worse lung function. Another study of 42 poorly controlled
asthmatics using 16S RNA phylochip sequencing on bronchial brushes found a greater bacterial
burden and diversity in asthma compared with health (Huang, Nelson et al. 2011). When these
subjects were treated with clarithromycin, greatest clinical response correlated with greater pre-
treatment bacterial airway diversity.
Timothy SC Hinks 1. Introduction
26
By what mechanisms might bacteria drive immunopathology in asthma? In the study by Huang,
100 taxa, mostly proteobacteria, were associated with bronchial hyper-responsiveness. Some of
these species might drive asthma through idiosyncratic mechanisms. For instance they
observed a Nitrosomonas species which can generate nitric oxide, and a Comamonadaceae
species which can degrade steroids (Huang, Nelson et al. 2011). Likewise, several studies have
provided evidence that nasal carriage of Staphylococcus may drive IgE mediated inflammation
through a super-antigenic effect (Bachert, Gevaert et al. 2003; Bachert, Gevaert et al. 2007),
whilst others have implicated Chlamydia pneumoniae infection in severe asthma(Black,
Scicchitano et al. 2000; ten Brinke, van Dissel et al. 2001; Biscione, Corne et al. 2004; Harju,
Leinonen et al. 2006).
In summary, study of the microbiome in asthma is as yet a nascent field, and to date few studies
have attempted to apply metagenomic techniques to correlate systematically the microbiome
with clinical and immunological metrics across a range of clinical phenotypes. Also, no studies
have attempted to integrate flow cytometric or microarray assessment of the innate or adaptive
immune system with unbiased analyses of the human lung metagenome.
Objectives
The primary goal of the work presented in this thesis was to elucidate the role of IL-17 and TH17
cells in relation to asthma severity and virus-induced asthma exacerbations relative to other key
CD4+ T lymphocyte subsets, namely TH1 and TH2 effector T-cells and regulatory
CD4+CD25+FOXP3+ Treg, as well as the less-researched cytotoxic T cells and the novel
mucosal associated invariant T cell subset (MAIT).
This goal was undertaken with the aim of improving characterisation of severe asthma versus
milder forms of asthma, thereby facilitating future progress in basic and applied research
(Anderson 2008). Moreover it was hoped this would deepen our understanding of the role of IL-
17 in the pathogenesis of asthma and host responses to respiratory virus infections, the
potential identification of new biomarkers for asthma phenotypes (Gibson 2007) and new targets
for pharmacological intervention. This goal was pursued through the investigation of two distinct
cohorts in two separate aims.
Aim 1
My initial aim outlined in my successful application for the Wellcome Trust Clinical Training
Fellowship, was to provide a detailed phenotypic characterisation of IL-17-producing cells in the
airways of mild, moderate and severe asthmatics.
My hypotheses were that:
i) The number of TH17 cells is raised in more severe forms of asthma.
Timothy SC Hinks 1. Introduction
27
ii) Dysregulation in the TH17:Treg balance would be associated with, and help define,
the severe neutrophilic asthma phenotype;
iii) Pulmonary TCRαβ+ TH17 cells would be a major primary cell source of IL-17 in
severe human asthma, while TCR+ T-cells would also contribute;
iv) TH17 cells would be enriched within the lung compartment and be predominantly
localised to the bronchial epithelium;
v) TH17 cells would be correlated with chronic virus infection (Wos, Sanak et al. 2008)
and airway bacterial colonisation(Simpson, Grissell et al. 2007; Simpson, Powell et
al. 2008).
Aim 2
Within Aim 2, I planned a longitudinal investigation into the dynamics of the TH17 response
during naturally occurring virally-induced exacerbations of asthma, to determine whether IL-17
is induced during naturally occurring asthma exacerbations, leading to neutrophilic infiltration.
In conjunction with a phase II, double-blind, randomised, placebo-controlled trial of inhaled
recombinant human (rh)IFN-β1b given at the onset of a common cold to asthmatic patients with
the aim of preventing/ameliorating an exacerbations, I undertook longitudinal follow-up of a well
characterised cohort of asthmatics with frequent exacerbations. This allowed me to study how
TH17 cells change during virus infections and associated asthma exacerbations as well as to
elucidate how treatment with IFN-β influences TH17 function.
My hypotheses were that:
i) Airway accumulation of TH17 cells would occur early in infection, leading to neutrophilia,
followed by a TH1 dominant response.
ii) Acute infection would be associated with a decrease in Treg frequency, which would be
more marked in asthma.
iii) Administration of inhaled rhIFN-β1b would inhibit the magnitude of the TH17 response to
viral infection measured in PBMC and airway samples.
Aim 3.
My aim was to perform the first analysis of MAIT cells within the human airways by observing
their frequencies in peripheral blood and in airway tissues in relationship to disease severity and
phenotype and to characterise their functional capabilities.
My hypotheses were that:
i) MAIT cells would be present in the human airways and concentrated in the airway
mucosa.
ii) MAIT cells would display pro-inflammatory effector function as judged by their
expression of cytokines.
Timothy SC Hinks 1. Introduction
28
iii) MAIT cell frequencies might be modulated by treatment with exogenous
corticosteroids.
iv) MAIT cell frequencies might vary following a seasonal pattern, possibly influenced by
variation in levels of vitamin D
Aim 4
My aim was to analyse the microbial metagenome in health and asthma to determine whether
asthma, particularly severe, steroid-resistant phenotypes, were associated with increased
detection of specific airway bacteria, or increased detection of respiratory viruses. I
hypothesised that:
i) Severe, steroid-resistant asthma would be associated with increased detection of
pathogenic airway bacteria including Haemophilus influenzae, Streptococcus
pneumoniae and Moraxella catarrhalis.
ii) Asthma may be associated with increased detection of viral genomes suggestive of
chronic viral infection or delayed viral clearance.
Timothy SC Hinks 2. Materials and Methods
29
CHAPTER 2
Materials and methods Nullius in verba 2
2 Motto of the Royal Society, chosen from Horace’s Epistles to signify the Fellows'
determination to establish facts via experiments. It may be translated ‘Take nobody's word
for it’.
Timothy SC Hinks 2. Materials and Methods
30
The objective of my thesis was to study the role of distinct T cell subsets in asthma in relation to
disease severity and virally-induced exacerbations. I therefore needed the following techniques: a
range of clinical measurements to obtain detailed clinical phenotyping of subjects, schemas by which
to classify participants according to a range of measures of asthma severity and techniques to obtain
clinical specimens of peripheral blood, sputum, bronchoalveolar lavage (BAL), bronchial epithelial
cells and bronchial biopsies for immunological and microbiological analysis. To enumerate T cells
accurately I needed to be able to maintain them in cell culture, stimulate them ex vivo and semi-
quantify cytokine production using intracellular cytokine staining. In order to characterise the
transcriptome of individual cell types I needed to sort cell populations, extract ribonucleic acid from
cells and analyse by quantitative polymerase chain reaction (qPCR) and/or microarray hybridisation.
In this chapter I will describe the methods used and will highlight those which I first needed to develop
in order to conduct the study.
Study design
This study comprised two components. The first component was a cross-sectional study in which
healthy volunteers and asthmatic subjects across a range of clinical phenotypes and disease severity
were compared by assessment of clinical and immunological parameters, undergoing phlebotomy,
sputum induction and bronchoscopy during periods of clinical stability (Figure 2.1). In addition
subgroups of subjects underwent repeated sampling after one week of inhaled or oral corticosteroids.
The second component was a longitudinal study in which I used the opportunity provided to me by a
clinical trial where IFN-β1α was studied for its effects on preventing or attenuating exacerbations
caused by upper respiratory tract infections (URTIs)(Figure 2.2). Subjects were sampled by
phlebotomy and sputum induction at baseline and at a further 7 time-points from the onset of the next
symptomatic URTI.
Timothy SC Hinks 2. Materials and Methods
31
Figure 2.1 Cross sectional study flow diagram
Flow diagram showing study design and recruitment for the cross sectional cohorts (Aim 1).
Timothy SC Hinks 2. Materials and Methods
32
Figure 2.2 Longitudinal study flow diagram
Flow diagram showing study design and recruitment for the longitudinal cohorts (Aim 2). Subjects
were recruited from SG005, a multicentre, multinational study, involving 26 sites, but samples gifted to
me all came from the Southampton site alone. 120 subjects were screened for eligibility. 47 subjects
developed exacerbations of asthma and were randomised to treatment, but of these samples were
available only from a subset for immediate (“fresh”) analysis of sputum and PBMC at 3 time-points
(n=13-14, unpaired samples at each time-point) or analysis of cryopreserved PBMC at 8 time-points
(n=26).
Timothy SC Hinks 2. Materials and Methods
33
Clinical measurements
Peak flow
Peak expiratory flow rates (PEFR) were measured using a European Standard mini wright peak flow
meter (EN 13826, Clement Clarke, Harlow, UK), with the subject standing. The highest of three
technically acceptable blows were recorded. European Community for Coal and Steel 1993 predicted
values were used(Quanjer, Tammeling et al. 1993).
PEFR Variability was measured using twice daily monitoring over 2 weeks and defined as (max PEFR
– min PEFR)/max PEFR x 100, expressed as a percentage according to British Thoracic Society
(BTS) guidelines (2008); ≥20% variability was considered significant.
Spirometry and reversibility
Spirometry was performed using Vitalograph dry wedge bellows (Vitalograph, Maids Moreton, UK)
and the following values calculated: forced expiratory volume in 1 second (FEV1), forced vital capacity
(FVC) and forced expiratory ratio (FER or FEV1/FVC). The best of 3 technically acceptable
manoeuvres were recorded where values were not ≥0.150L different between the largest and the next
largest FEV1/FVC results and were within 5% of each other (whichever was greater), according to
European Respiratory Society (ERS) guidelines (Miller, Hankinson et al. 2005).
Bronchodilator reversibility was tested using spirometry before and 12-15 minutes after administration
of salbutamol 400 mcg using a pressurised metered dose inhaler (pMDI) with a Volumatic device
(Allen and Hanbury’s, UK) or 2.5mg salbutamol via oxygen driven nebuliser, according to ERS
guidelines (Miller, Hankinson et al. 2005). Reversibility was defined as (post bronchodilator FEV1 - Pre
bronchodilator FEV1)/Pre bronchodilator FEV1 x 100, with a 12% increase considered
significant(Goldstein, Veza et al. 2001), according to BTS guidelines(2008).
Home monitoring
During the longitudinal study home monitoring of PEFR, FEV1 and FVC were performed using a Mini-
Wright Digital (Clement Clark, Harlow, UK) and downloaded intermittently using the manufacturer’s
software (MWD Soft 1.73).
TLCO
The transfer factor of lung carbon monoxide (TLCO) was measured using a single breath diffusion
method with a Morgan CPL device (Morgan Scientific, Haverhill, MA, USA) and expressed in
mmolmin-1kPa-1 using the Jones-Mead method with a 10 second (+/- 2 seconds) breath hold time
using the machine’s standard haemoglobin value and using the European predicted values the mean
of the two repeatable TLCO values, within 10% of each other. The TLCO was obtained from the
product of the two primary measurements: transfer coefficient (Krogh’s KCO) and alveolar volume (VA)
according to Krogh (Krogh 1915).
Timothy SC Hinks 2. Materials and Methods
34
Exhaled nitric oxide
Exhaled nitric oxide (eNO or FENO) was measured with a single breath at a flow rate of 50 mL/s using
a Niox Mino device (Aerocrine Inc, Princeton, NJ, USA), with an upper limit of normal for adults of <25
ppb.(Taylor, Pijnenburg et al. 2006)
Methacholine challenge testing of airway hyper-responsiveness
Airway hyper-responsiveness (AHR or bronchial hyper-reactivity (BHR)) was measured in all patients
except those with severe asthma, or those with recent historical data available. Spirometry and dosing
were performed with a Viasys APS system (CareFusion UK, Basingstoke, UK) using a fixed
concentration of 32mg/ml methacholine, over the dose range 0.0256 mg – 1.434 mg. Results were
reported as the non-cumulative provocative dose of methacholine causing a 20% fall in the FEV1
(‘non-cumulative PD20’), which is automatically calculated using logarithmic interpolation (Schulze,
Rosewich et al. 2009).
Historical data were accepted if performed within 1 year for asthmatics or 5 years for healthy controls.
In six instances provocative concentration (PC20) data only were available rather than PD20 data and
were converted to an approximate equivalent PD20 value by linear regression. In six healthy
individuals negative bronchial challenges had been performed with histamine and in each instance a
20% fall in FEV1 was not achieved with >8mg histamine.
Interpretation: methacholine challenge testing has a specificity of 90 to 95% and a sensitivity of 60-
100% for detecting physician-diagnosed asthma (Soysal, Bahceciler et al. 2008). PD20 values were
interpreted according to the ATS categorisation of bronchial responsiveness (Schulze, Rosewich et al.
2009) as follows:
Table 2.0
Non-cumulative methacholine
PD20 (mg)
Interpretation
>1.0 Normal bronchial response
0.6-1.0 Borderline BHR
0.3-0.6 Mild BHR
<0.3 Moderate-severe BHR
Skin prick allergen testing
Skin prick allergen testing (SPT), which is a functional assay of specific IgE responses on mast cells
in the skin and is associated with type 1 hypersensitivity (Gould, Sutton et al. 2003), was measured
with a panel of common aeroallergens. Allergen solutions used were aspergillus fumigatus, candida
albicans, mixed grass pollen, dermatophagoides pteronyssinus, dermatophagoides farina
Timothy SC Hinks 2. Materials and Methods
35
(respectively European and American house dust mites) feathers, cat and dog (Allergopharma,
Reinbek, Germany). Some historical data contained results for mixed tree pollen and for alternaria
tenuis. Positive and negative controls were histamine solution and carrier solution respectively. A
small drop of each solution was placed on the skin of the volar aspect of the lower forearm after it had
been cleaned with water and dried. Disposable sterile lancets (Allergopharma) were used to break the
dermis under each drop in turn at a 90 ̊ angle and the diameter of the weal measured in two
perpendicular directions after 15 minutes. A positive result was recorded if the weal was 3mm > result
of negative control.
Study populations
The study was approved by the National Research Ethics Service Committee South Central –
Southampton B ethics committee (Ref 10/H0504/2). Subjects were recruited who were willing to
participate and who met the following specific inclusion criteria, dependent on cohort (see Figure 2.1):
Cross sectional study (Aim 1)
General inclusion criteria
Able to provide written informed consent.
Aged 18-70.
Vital signs; at the discretion of the Investigator.
Motivation to complete all of the study visits and ability to communicate well with the
investigator and be capable of understanding the nature of the research and its treatment
including risks and benefits.
Inclusion criteria for healthy, non-atopic, non-asthmatic controls
All of the general inclusion criteria above.
Absence of atopy on skin prick allergen testing.
Absence of bronchial hyper-responsiveness: methacholine PD20 >1.0 mg.
Not a current smoker.
Inclusion criteria for steroid-responsive mild or moderate asthmatic subjects
All of the general inclusion criteria above.
A clinical diagnosis of asthma.
Presence of atopy on skin prick allergen testing.
Best described as mild or moderate asthma by the definition in table 2.2.
Inclusion Criteria for Severe Asthmatic Subjects
All of the general inclusion criteria above.
A clinical diagnosis of asthma.
Best described as “severe asthma” by the definition in table 2.2.
Exclusion criteria
Unable to provide written informed consent.
Timothy SC Hinks 2. Materials and Methods
36
Pregnancy either current or planned over the duration of the study.
Prisoners.
Children under age 16.
Lung disease other than asthma.
Additional older healthy controls
Subjects aged 40-70 years, able to provide written informed consent, with no history of atopy, asthma
or other lung disease.
Longitudinal study (Aim 2a)
Samples of blood and sputum were gifted to me by Synairgen Research Ltd from the trial SG005 “A
randomised, double-blind, placebo-controlled Phase II study, comparing the efficacy and safety of
inhaled SNG001 to placebo administered to asthmatic subjects after the onset of a respiratory viral
infection for the prevention or attenuation of asthma symptoms caused by respiratory viruses”
(NCT01126177)(See Figure 2.2) Full inclusion criteria are published available from the U.S. National
Institutes of Health (SynairgenResearchLtd 2012). Briefly, all subjects were aged 18-65 years with a
history of asthma for at least 2 years and confirmed by bronchodilator reversibility or bronchial hyper-
responsiveness or an exacerbation requiring medical review or hospital admission and who were
treated with regular inhaled corticosteroids and had a previous history of virus-induced exacerbations.
In addition to these criteria for baseline visits, to be randomised for inclusion in the exacerbation study
subjects needed to have been experiencing respiratory virus symptoms within the previous 24 hours,
being either cold symptoms (a blocked or runny nose, or sore throat) or influenza-like illness (fever
>37.8 ̊C plus two of headache, cough, sore throat or myalgia).
Clinical classification
Disease heterogeneity is increasingly recognised in asthma, with a recognition that there is a need to
elucidate distinct disease endotypes(Anderson 2008). Accordingly I have attempted to analyse data
using a variety of dimensions including continuous variables such as lung function (e.g. FEV1) or
symptom scores (e.g. ACQ) or treatment (e.g. step on the BTS treatment algorithm, see Figure 2.3),
or level of asthma control, (see table 2.1) or according to inflammatory subtype based on sputum cell
differentials (see Definitions of inflammatory subtypes).
Asthma control questionnaire
Subjects were phenotyped along a scale of disease control according to the well validated asthma
control questionnaire (ACQ) (Juniper, O'Byrne et al. 2000) which combines subjective self-
assessments of disease control with objective assessment of lung function into a global score. in a
study of 1323 individuals Juniper et al have shown that the crossover point between 'well-controlled'
and 'not well-controlled' is close to 1.00 on the ACQ, whilst a cut-point of 1.50 can be used to
confidently predict inadequately controlled asthma with a positive predictive value of 0.88 (Juniper,
Bousquet et al. 2006).
Timothy SC Hinks 2. Materials and Methods
37
Figure 2.3 BTS treatment algorithm
British Thoracic Society treatment algorithm for asthma in adults. From (2008) p iv42 “Summary of
stepwise management in adults”.
Timothy SC Hinks 2. Materials and Methods
38
ASTHMA CONTROL QUESTIONNAIRE
The ACQ consisted of the following questions and instructions:
Circle the number of the response that best describes how you have been during the past week.
1. On average, during the past week, how often were you woken by your asthma during the night?
0. Never 1. Hardly ever 2. A few times 3. Several times 4. A great many times 6. Unable to sleep because of asthma
2. On average, during the past week, how bad were your asthma symptoms when you woke up in the morning?
0. No symptoms 1. Very mild symptoms 2. Mild symptoms 3. Moderate symptoms 4. Quite severe symptoms 5. Severe symptoms 6. Very severe symptoms
3. In general, during the past week, how limited were you in your activities because of your asthma.
0. Not limited at all 1. Very slightly limited 2. Slightly limited 3. Moderately limited 4. Very limited 5. Extremely limited 6. Totally limited
4. In general, during the past week, how much shortness of breath did you experience because of your asthma?
0. None 1. Very slightly limited 2. A little 3. A moderate amount 4. Quite a lot 5. A great deal 6. A very great deal
5. In general, during the past week, how much of the time did you wheeze?
0. Not at all 1. Hardly any of the time 2. A little of the time 3. A moderate amount of the time 4. A lot of the time 5. Most of the time 6. All of the time
6. On average, during the past week, how many puffs of short-acting bronchodilator (e.g.Ventolin) have you used each day?
0. None 1. 1-2 puffs most days 2. 3-4 puffs most days 3. 5-8 puffs most days 4. 9-12 puffs most days 5. 13-16 puffs most days 6. More than 16 puffs most days
Point 7 of the ACQ is completed by the investigator based on pre-bronchodilator FEV1. 0, 95% predicted; 1, 95–90% 2, 89–80%; 3, 79–70%; 4, 69–60%; 5. 59–50%; 6. 50%
Timothy SC Hinks 2. Materials and Methods
39
Table 2.1 Levels of asthma control, GINA ((GINA) 2010)
Characteristic Controlled Partly controlled Uncontrolled
Daytime symptoms None (twice or
less/week)
More than twice/week Three or more
features of partly
controlled asthma
present in any week Limitation of activities None Any
Nocturnal symptoms /
awakening
None Any
Need for reliever /
rescue treatment
Non (twice or
less/week)
More than twice/week
Lung function (PEF or
FEV1)
Normal <80% predicted or
personal best (if
known)
Exacerbations None One or more/year One in any week
I have also attempted to use cluster analysis to identify distinct endotypes based on a more objective
synthesis of all these distinct, interacting factors. I have also stratified subjects into distinct clinical
phenotypes by considering asthma as a single disease along a continuous spectrum of severity,
according to the following scheme which is based on a global assessment of disease severity and
largely derives from the GINA classification of asthma severity ((GINA) 2010).
Timothy SC Hinks 2. Materials and Methods
40
Table 2.2 Definitions of asthma severity used in this project
Mild intermittent asthma
Mild
ast
hma
Symptoms once a week Nocturnal symptoms not more than twice a month • FEV1 or PEF ≥ 80% predicted Treatment: • Salbutamol as needed only PD20 is needed for these people to ensure correct diagnosis. If borderline, the result has to be interpreted in context with the history.
Mild persistent asthma
Symptoms > once a week but < once a day Nocturnal symptoms < twice a month • FEV1 or PEF ≥ 80% predicted Treatment: • Salbutamol as needed only
Moderate persistent but well controlled asthma
Mod
erat
e as
thm
a
Symptoms <3x/week Nocturnal symptoms <twice a month • FEV1 or PEF ≥ 80% of predicted or of patient’s best Treatment: • Salbutamol as needed only • Low-dose (<800 µg BDP equivalent) inhaled steroids • +/- Long acting beta-2-agonist
Moderate persistent but not well controlled asthma
Symptoms >3x/week Nocturnal symptoms >twice a month (some may not have nocturnal symptoms) • FEV1 or PEF <80% of predicted or patient’s best Treatment: • Salbutamol as needed only • Low-dose inhaled steroids • +/- Long acting beta-2-agonist
Severe asthma
Sev
ere
asth
ma
Symptoms daily Nocturnal symptoms >once a week Daily use of inhaled short-acting ß2-agonist • FEV1 or PEF <80% of predicted or patient’s best Treatment:
High-dose (at least 800 µg of BDP equivalent) inhaled steroids
Long acting beta-2-agonist
+/- frequent or continuous oral steroids
Notes: where patients do not fit neatly into any category they are considered on an individual basis to
make the best possible fit. Not all criteria have to be fulfilled for any of these categories.
Timothy SC Hinks 2. Materials and Methods
41
Phlebotomy
Serum
Samples for serum were obtained using the 21 gauge BD Vacutainer® Safety-LokTM blood collection
set (367282, BD, Plymouth, UK) into 10 ml BD Vacutainer® serum tubes (367895, BD) and allowed to
clot in an upright position at room temperature (RT)RT for ≥30 minutes. Samples were then
centrifuged at 1653g for 10 minutes and supernatants aspirated and frozen in 1 ml aliquots at -80 ̊C
until required.
Full blood count
Samples for full blood count were taken into 3ml EDTA tubes (367838, BD) and processed in the
National Health Service laboratory by automated cytometers.
Peripheral blood mononuclear cell preparation
Blood for isolation of peripheral blood mononuclear cells (PBMC) was obtained into 6 ml lithium
heparin tubes (367885, BD). Dulbecco’s phosphate buffered saline with no Mg2+ or Ca2+ (PBS,
D8537, Sigma-Aldrich, Gillingham, UK) was warmed and added to the heparinised blood in a 1:1 ratio
in 50ml falcons. This mixture was carefully layered over LymphoprepTM (Nycomed/Axis-Shield PoC,
Rodeløkka, Norway) or Ficoll-PaqueTM (17-1440-03, GE Healthcare, Uppsala, Sweden) in the ratio
20ml lymphoprep:30ml blood/PBS mixture) and centrifuged at 800g for 30 mins at 20°C with the
brake off. The buffy coat layer was aspirated using a pastette, into a fresh 50ml falcon and washed
twice with 50ml with PBS and centrifugation at 400g for 5 mins. 10μL of cells were removed for
counting in trypan blue (T8154, Sigma) with a Neubauer haemocytometer (Marienfeld, Lauda-
Königshofen, Germany). Cells were then resuspended at appropriate concentrations in culture
medium or PBS as required.
Cell preparation tubes
To minimise lab processing time, blood samples from the longitudinal study were taken directly into
cell preparation tubes (362780, BD) which contain sodium heparin and Ficoll. Tubes were inverted 8-
10 times, then centrifuged at 1650g for 20 minutes with the brake off. Using a pastette the top half of
plasma layer was aspirated without disturbing the mononuclear cell and platelet layer and the
mononuclear cell layer transferred to a fresh 15ml falcon. Samples were washed twice with 15ml
warmed Roswell Park Memorial Institute medium 1640 without L-glutamine or phenol red (RPMI,
R7509, Sigma-Aldrich, Gillingham, UK or Lonza/Biowhittaker, Basel, Switzerland) and with
centrifugation at 400g for 5 mins.
Nasal lavage
Nasal lavage was performed during the longitudinal study using reusable metal nasal olives and a 5ml
syringe. The subject was asked first to blow their nose, then sit upright with head slightly forward
Timothy SC Hinks 2. Materials and Methods
42
whilst 2.5ml of normal saline was gently introduced and withdrawn 5 times, before being collected
with a funnel into a sterile universal container and repeating in the other nostril with a further 2.5ml.
Sputum induction
Sputum induction protocol
Sputum induction was performed using 4.5% hypertonic saline, except in higher risk patients,
according to an established protocol (Djukanovic, Sterk et al. 2002). Baseline spirometry and
salbutamol reversibility testing were performed, then subjects inhaled 4.5% hypertonic saline for 5
minutes, using an ultrasonic device (DeVilbiss UltraNeb, Tipton, UK) in an environmental chamber
(Protex, Halifax, UK). Subjects were advised to maintain normal tidal breathing and wore a nose clip.
After 5 minutes FEV1 were recorded and any sputum expectorated, before the procedure was
repeated a further three times, unless an adequate sample were obtained or FEV1 fell by 20%. The
chance of a successful sputum induction was maximised by encouraging good hydration and using
the technique of autogenic postural drainage.
An alternative protocol for high risk subjects was used if post-bronchodilator FEV1 was < 60% of
predicted or <1.5L or there was a history of severe asthma, or highly reactive airways. In this protocol
subjects received 0.9% saline for 0.5, 1, 5 mins then 3% saline for 0.5, 1, 2 mins, then 4.5% saline for
0.5, 1, 2, 4, 8 minutes(Djukanovic, Sterk et al. 2002).
Sputum processing
Sputum samples were kept in a petri dish on ice and forceps used to select mucus plugs for transfer
into a falcon tube. Samples were weighed and diluted with four volumes of the reducing agent 1,4
dithioerythritol (DTE) in solution (5mM DTE(Sigma), 28.8mM HEPES buffer (Lonza), 30mM NaCl) and
placed on a bench rocker for 30 mins, with intermittent homgenisation with a pastette. Mucus was
removed with a 100μm filter to remove mucus and filtrate centrifuged at 400g for 10 mins to pellet
cells. Aliquots of supernatant were ultracentrifuged at 12000g for 5 minutes to precipitate bacteria and
then frozen at -80 ̊C and cells resuspended in appropriate medium.
Preparation of cytospins
Twinfrost glass microscope slides (CellPath, Newtown, UK) were pre-coated with poly-L-lysine
(Sigma) by immersion in 0.01% (w/v) poly-L-lysine in distilled water for 5 mins, then air dried at RT
overnight. Slides were labelled and placed in a cage with a cytofunnel and hole-punched filter paper
in a Shandon Cytospin centrifuge. Sputum or bronchoalveolar lavage cells were resuspended at a
concentration of 1x106 cells/ml in PBS or culture medium and 70 μL per slide transferred to the
cytofunnel, then centrifuged at 450 revolutions per minute (rpm) for 6 minutes.
Slides were then air dried for at least 24 hours and stained with a rapid Romanowsky
stain(Jorundsson, Lumsden et al. 1999) (Raymond Lamb, Eastbourne, UK): slides were immersed in
methanol fixative for 30 seconds, blotted, stained for 30 seconds in eosin, blotted, stained for 1
Timothy SC Hinks 2. Materials and Methods
43
minute in methylene blue, then rinsed in running tap water and allow to air dry. Once dry slides were
mounted with Pertex (Cell Path) and a coverslip and 400 cells counted manually, recording
eosinophils, neutrophils, macrophages/monocytes, lymphocytes and columnar epithelial cells. In
addition numbers of squamous cells were noted but not included in the differential. A count of more
than 30% squamous cells was suggestive of significant upper airway contamination
(Hadjicharalambous, Dent et al. 2004; Singh, Edwards et al. 2010). Cytospin staining and differential
counts were performed by my colleague Jon Ward.
Definitions of inflammatory subtypes
For the purposes of this work the following definitions were used for subtypes of asthma.
Inflammatory subtypes Definition (based on sputum cell differential)
Neutrophilic asthma ≥61% neutrophils
Eosinophilic asthma ≥3% eosinophils
Mixed granulocytic asthma ≥3% eosinophils and ≥61% neutrophils
Paucicellular asthma <3% eosinophils and <61% neutrophils
This definition of neutrophilic asthma is widely accepted (Simpson, Scott et al. 2006; Haldar and
Pavord 2007; Simpson, Grissell et al. 2007; Cowan, Cowan et al. 2010), as is the division of asthma
into these four inflammatory subtypes according to sputum cytospin cell differentials. However there is
less agreement over the definition of sputum eosinophilia, with different authors choosing cut-offs of
1% (Simpson, Scott et al. 2006), 2% (Jayaram, Pizzichini et al. 2006), and 3% (Green, Brightling et al.
2002; Green and Pavord 2012). I have chosen to use a 3% cut-off because the normal range of
sputum eosinophils in adults and children is <2.5% (Spanevello, Confalonieri et al. 2000; Kips, Inman
et al. 2002) and because it has been shown that a 3% cut-off identifies individuals with corticosteroid-
responsive asthma (Pavord, Brightling et al. 1999; Green, Brightling et al. 2002).
Bronchoscopy
Bronchoscopic technique
Bronchoscopies were performed using Pentax video colour CCD flexible bronchoscope (Pentax UK,
Slough, UK) in a purpose-built research endoscopy suite with the assistance of at least 2 nurses and
a laboratory technician and in accordance with the BTS guidelines current at the time (2001) and with
established research protocols(Jarjour, Peters et al. 1998). Briefly, subjects who had been starved for
at least 4 hours underwent routine physical examination and routine measurement of vital signs,
spirometry and reversibility(1991), including premedication with 2.5 mg of nebulised salbutamol. An
intravenous cannula (Biovalve, Vygon, Swindon UK) was inserted and the procedure performed under
light sedation with alfentanyl 0-1000 mcg and / or midazolam 0-10 mg with continuous pulse oximetry.
The analgesia and suppression of gag and cough reflexes was achieved with 6-8 sprays (60-80 mg)
Timothy SC Hinks 2. Materials and Methods
44
of 10% lidocaine orally, 2-5 ml of Instillagel (CliniMed, High Wycombe, UK) nasally, 6 ml of 2%
lidocaine to the vocal cords and 5-10 ml of 1% lidocaine to the bronchial tree.
Bronchial brushings were obtained from right and left lower and middle lobe 1st-3rd order bronchi by
gentle brushing using sterile 2 mm disposable cytology brushes (BC-202D-2010, Olympus UK,
Southend-on-Sea, UK) and samples gently agitated in 5ml of ice-cold PBS. Brushings were taken
prior to other samples to minimise contamination with blood which would decrease the cellular purity
of samples and can impair growth of primary bronchial epithelial cells in vitro.
Bronchoalveolar lavage (BAL) was performed in the right upper lobe, usually in the posterior segment,
as the anatomy favours good recovery volumes. Where possible sample contamination from the
bronchoscopes was minimised by taking some or all of the BAL through a sterile, 2mm diameter
disposable protected catheter (Combicath, ConMed Linvatech, Swindon, UK). After anaesthesia of
the relevant lobe the catheter was passed through the bronchoscope and the inner catheter fully
inserted then removed, to expel the wax seal. The outer catheter was then placed into the bronchus
during the wedge and the lavage taken through the outer catheter using 6 x 20 ml warmed, sterile
normal saline, injecting and aspirating with 20 ml syringes. The first 1 ml of BAL recovered was
dispensed into a sterile eppendorf and stored at -80 ̊C for later microbiological analysis. In some
instances the complete lavage could not be performed through the Combicath and the procedure was
converted to a standard technique at this stage.
Up to 10 bronchial biopsies were taken from 1st-4th order carinae using 1.8 mm alligator cup biopsy
forceps (100503, ConMed) and samples gently transferred to ice-cold RPMI.
After the procedure subjects were observed for at least 60 minutes, before having their swallow tested
and repeat spirometry performed.
See Figure 2.4 for overview of sample processing.
Timothy SC Hinks 2. Materials and Methods
45
Figure 2.4 Sample processing
Flow diagram showing the processing of samples taken at a bronchoscopy visit in the cross sectional
study. Not shown are additional sputum processing steps performed at a separate visit which
occurred ≥1 week before or 4 weeks after the bronchoscopy visit.
Processing of BAL
Bronchoalveolar lavage was transferred to the laboratory on ice then, except the aliquot kept for
microbiological analysis, was strained through 100μm cell strainer and centrifuged at 400g, 5 mins, at
4°C. Aliquots of supernatant were ultracentrifuged at 12000g for 5 minutes to precipitate bacteria and
then frozen at -80 ̊C and cells resuspended in appropriate medium. If samples were heavily blood-
stained red blood cells were haemolysed by resuspending in 4.5ml sterile water for 30 seconds,
followed immediately by addition of 0.5ml 10x Hank’s balanced salt solution (HBSS, Gibco) and made
up to 50 ml with PBS, before centrifuging again at 400g for 5 mins. Next, cells were resuspended in
1ml AIM V, 10 μL removed for viability counts and the concentration adjusted to 1x106 cells/ml for
culture or immediate staining for flow cytometry. 2x70 μL of cell suspension were removed to produce
cytospins as previously described.
Processing of bronchial biopsies
Biopsies were transferred to the laboratory on ice, washed in fresh RPMI to remove residual blood,
weighed in pre-weighed eppendorf tubes and transferred using pastettes either into pre-warmed AIM
Timothy SC Hinks 2. Materials and Methods
46
V medium for overnight culture or pre-warmed collagenase solution for immediate staining for flow
cytometry.
Collagenase digestion of biopsies
Biopsies were dispersed in 2.5ml of type 1 collagenase from Clostridium histolyticum, (1mg/ml C0130,
Sigma) reconstituted in RPMI for 1hr at 37°C with magnetic stirring. Cells were then passed through a
70μm filter, centrifuged at 400g for 5 mins and resuspended in PBS for immediate staining.
This method was originally described for isolation of T cells from adipose tissue, where it was shown
that type 1 collagenase is superior to other enzyme preparations - collagenase IV and liberases which
are a blend of collagenase I and II with a neutral protease or dispase - with respect to cell viability,
yield and preservation of cell surface markers (Hagman, Kuzma et al. 2012). These authors also
showed that the optimal time for digest was 60-75 mins because longer digests caused significantly
more loss of all surface markers tested(Hagman, Kuzma et al. 2012). This protocol has subsequently
been used in our group for dispersion bronchial biopsies (Vijayanand, Seumois et al. 2007; Ganesan
2010) but is complicated by cleavage of the surface CD4 co-receptor (Figure 2.5)(Hagman, Kuzma et
al. 2012). For this reason it was necessary in the case of bronchial biopsy derived samples to identify
T helper cells by negatively selecting on CD8.
Timothy SC Hinks 2. Materials and Methods
47
Figure 2.5 Cleavage of CD4 by collagenase dispersion
Surface staining for CD4 and CD8 expression on T cells obtained by either bronchoalveolar lavage
(A), or by collagenase dispersion of bronchial biopsies (B), reveals relative preservation of CD8 but
dramatic loss of CD4 staining due to cleavage of CD4 by collagenase. Histograms show CD4
brightness, with complete loss of the bimodal distribution of CD4 expression in the case of biopsies.
(MFI, Mean Fluorescence Intensity).
Timothy SC Hinks 2. Materials and Methods
48
Culture media
The following culture media were used for all other experimental work:
RPMI
Roswell Park Memorial Institute medium 1640 without L-glutamine or phenol red (R7509, Sigma or
Lonza/Biowhittaker)
Complete serum free medium (AIM V)
AIM V® Medium (Gibco, Life Technologies, Paisley, UK) supplemented with:
0.5 μg/ml Fungizone (Amphotericin B with sodium deoxycholate, Gibco)
2 mM L-glutamine (Gibco)
1 mM sodium pyruvate (Gibco)
100 µg/ml streptomycin (Gibco)
100 U/ml penicillin (Gibco)
0.004% (v/v) 2-mercaptoethanol (β-ME)(Stratagene)
RN10 culture medium with 10% human serum
RPMI 1640 (Sigma or Lonza/Biowhittaker) supplemented with:
2 mM L-glutamine (Gibco)
1 mM sodium pyruvate (Gibco)
100 µg/ml streptomycin (Gibco)
100 U/ml penicillin (Gibco)
50 ml human AB serum (heat inactivated at 56º C for 30 min in water bath)(Sigma)
T cell growth medium
RPMI 1640 supplemented with:
10% (v/v) foetal calf serum
2 mM L-glutamine (GlutaMAX™, GIBCO)
2 mM sodium pyruvate (Gibco)
100 µg/ml streptomycin (Gibco)
100 U/ml penicillin (Gibco)
MEM essential amino acids (M5550, Sigma)
Non-essential amino acids (M7145, Sigma)
400 U/ml rh-IL2 (Proleukin, Prometheus Laboratories, San Diego, CA, USA)
10 ng/ml rhIL-7 (Immunotools Gmbh, Friesoythe, Germany)
10 ng/ml rhIL-15 (Immunotools)
T cell sorting medium
T cell growth medium, as above, supplemented with
2% human serum (Sigma)
Timothy SC Hinks 2. Materials and Methods
49
0.1 mg/ml Kanamycin
50 μM β-mercaptoethanol
At the time of sorting and intermittently thereafter cells were additionally stimulated with
phytohaemagglutanin (PHA-p, Sigma) at 1 μg/ml.
Magnetic-activated cell sorting (MACS) Buffer
2 mM ethylenediamine tetraacetic acid (EDTA, Fluka BioUltra, Sigma) 0.5% (w/v) bovine serum
albumin (BSA, Sigma) 0.22 μm filter-sterilised in PBS.
Cryopreservation of cells
PBMC from the longitudinal cohort, as well as cloned cell lines were cryopreserved for long term
storage. Fresh cryopreservation solution was made on the day of use (Simione 1998) comprising heat
inactivated human serum albumin (Sera Laboratories, Haywards Heath, UK, inactivated at 56º C for
30 min in a water bath) with 10% dimethyl sulphoxide (DMSO, Sigma) and cooled for at least 15
minutes on ice prior to use, as this is an exothermic reaction.
Supernatant was poured off and cells resuspended in residual volume of RPMI by flicking the tubes.
They were then resuspended in pre-labelled cryovials on ice, (H.A.N.C. 2011) to a final concentration
of 2-10x106 cells/ml, in 2 ml ice-cold cryopreservation solution, added drop-wise over at least 2 mins
to minimise osmotic shock(Jeurink, Vissers et al. 2008), with intermittent gentle shaking. The cryovials
were then transferred immediately to a “Mr Frosty” 5100 Cryo 1°C Freezing Container (Nalgene,
Thermo Fisher Scientific, Langenselbold, Germany) containing isopropanol which was pre-warmed to
RT. This was placed immediately into a -80 ̊C freezer which allows approximately 20 minutes for
DMSO to equilibrate with the cells whilst they cool at 1 ̊C/min(Simione 1998). The following day
samples were transferred to liquid nitrogen (-196 °C) for long term storage.
Human serum albumin was used in preference to foetal calf serum or human AB serum because,
where freezing protocols have been compared directly, this provided the best cell viability and
retention of lymphocyte function after cryopreservation (Disis, dela Rosa et al. 2006). In addition
human serum albumin is less likely to activate cells in functional assays and gives better results in
FOXP3 staining (Ganesan 2010). Other factors associated with improved lymphocyte function after
cryopreservation include freezing cells at a concentration of 2-4x106cells/ml, as lower yields are
obtained at lower concentrations((ECACC) ; Simione 1998), gentle handling during cell harvesting
and concentration procedures; avoiding vigorous pipetting and high-speed centrifugation and
ensuring that cells are defrosted at exactly 37 ̊C (Disis, dela Rosa et al. 2006). Factors which have
been shown not to affect cell viability include using cell preparation tubes or Ficoll Paque (Ruitenberg,
Mulder et al. 2006), transferring cells to liquid nitrogen or keeping on dry ice (-78.5 ̊C) for 3 days,
thawing in 15ml tube or 50ml tube, centrifuging for 5 or 10mins and at 1200rpm or 1500 rpm during
washing or freezing at 10x106 cells/ml or 30x106 cells/ ml(Disis, dela Rosa et al. 2006).
Timothy SC Hinks 2. Materials and Methods
50
Thawing cryopreserved cells
Cryovials were removed from liquid nitrogen and immediately warmed at exactly 37ºC in a water bath.
As soon as the last ice crystals had disappeared the suspension was pipetted drop-wise into 10ml of
pre-warmed RN10 medium, centrifuged immediately at 400g for 5 mins, the supernatant removed and
cells resuspended in 1ml of RN10 to approximately 1x106 cells /ml. Cells were then rested by
incubation overnight in single 0.5ml wells at 37°C, 5% CO2 to allow them to adapt to culture conditions
(Maecker, Moon et al. 2005; Jeurink, Vissers et al. 2008) prior to restimulation for 5 hours in RN10
rather than serum free media.
Enzyme linked immunosorbent assay (ELISA)
Measurement of total immunoglobulin E (IgE)
Frozen serum aliquots were defrosted at RT and transferred into a 96-well round bottomed plate, then
assayed using a Platinum ELISA kit (BMS2097, eBioscience, Hatfield, UK) according to the
manufacturer’s instructions. Briefly 96-well microwell plates pre-coated with monoclonal antibody to
anti-human IgE were washed twice with 400 μL of wash buffer (0.05% (v/v) Tween 20 (P-1379,
Sigma) in PBS (X6571D, Oxoid, Basingstoke, UK)(pH 7.3)). Duplicate standard curves were prepared
in 100 μL of assay buffer (0.05% (v/v) Tween 20 in PBS with 10% BSA), using doubling dilutions of
standard IgE protein (500 ng/ml to 7.8 ng/ml). 10 μL of samples were transferred from the round
bottom plate and added in single wells to 90 μL of assay buffer in the assay plate. Duplicate blank
wells contained 100μL of assay buffer only. Fifty microliters of horse radish peroxidase (HRP)-
conjugated anti-human IgE monoclonal antibody were added to all wells and the plates covered with
adhesive film and incubated for 60 minutes at RT with vigorous shaking (150 rpm). Wells were then
emptied and washed twice with 400 μL of wash buffer and then 100 μL of tetramethyl-benzidine
(TMB) substrate solution were added to all wells. The plates were covered and incubated for 25
minutes in the dark. 100 μL of stop solution (1M phosphoric acid) was added to each well and 450 nm
absorbance measured on a plate reader (Multiskan Ascent, Agilent Technologies, Wokingham, UK).
Concentrations were determined from a standard curve using 5-parameter curve fit. Concentrations in
ng/ml were converted to international units (IU)/ml according to the manufacturer’s comparison with
the WHO Reference Serum (NIBSC code: 75/502, 1 IU/ml corresponds to 2.44 ng/ml).
Measurement of IL-17
Measurement of the concentration of IL17-A protein was attempted in supernatants of sputum, BAL
and culture-conditioned media from allergen stimulated biopsies and virally infected parenchymal
explants, by ELISA using pre-coated plates (88-7976, eBioscience, clone eBio64CAP17) according to
the manufacturer’s instructions. Briefly IL17-A standard was diluted in 100 μl assay buffer in duplicate
2-fold serial dilutions over the range 3.9-500 pg/ml. Samples were diluted 1:1 (BAL and culture-
conditioned media) or 1:3 (sputum) with assay buffer and 100 μl added to wells in duplicate, plates
sealed and incubated at RT for 2 hours. Next wells were aspirated and washed 5 times with 250 μl /
Timothy SC Hinks 2. Materials and Methods
51
well wash buffer, blotted and incubated for 1 hour at RT with 100 μl/well of detection antibody
(eBio64DEC17). Plates were again aspirated and washed 5 times before incubating for 30 mins at RT
with 100 μl/well of avidin-horseradish peroxidase, then aspirating and washing a further 7 times,
before incubating with 100 μl/well of TMB substrate solution at RT for 15 minutes. The reaction was
stopped with 50 μl of stop solution (5.7% phosphoric acid) per well and the plates read at 450 nm.
Meso-Scale Discovery platform
The Meso Scale Discovery multi-array assay platform (MSD) was selected as it is robust and, can be
multiplexed with small sample volumes and has have a lower limit of detection than ELISA or
Luminex, especially as for IL-17A which quotes a lower limit of detection in serum of 0.4pg/ml.
Samples were analysed by MSD according to the manufacturer’s instructions using the IL17
ultrasensitive single-plex assay (K151ATC-2, Meso Scale Discovery, Gaithersburg, USA) for IL17-A
and the TH1/2 7-cytokine multiplex assay (K15011C-2 Meso Scale Discovery) for IFN-γ, IL-10, IL-12
p70, IL-13, IL-2, IL-4, IL-5, with standard curves made up both in Diluent 2 and in DTE/Diluent 2 1:1
mix. The standard protocol was modified in discussion with the manufacturer according to the sample
type.
BAL was first concentrated and a final concentration of approximately 1% bovine serum albumin
(BSA) was achieved by addition of 20 μL of 10% BSA to the samples prior to concentration, to act as
a carrier protein. 6 x 1ml aliquots of each sample were transferred to Vivaspin 6 centrifugal
concentrators (Sartorius Stedim Biotech GmbH, Göettingen, Germany) and centrifuged at 4 ̊C at
1570-2100g for 4-5 hours until the dead-stop volume was approached, to achieve a dilution of 11.3-50
(median 44.7) fold. An additional sample of pooled BAL samples was spiked with 50 pg/ml of cytokine
standards. The standard curve for BAL samples was diluted in PBS with 1% (w/v) BSA.
Sputum samples and the DTE standard curve were diluted 1:1 in diluent 2. This was a compromise
between the denaturing effect of the reducing agent and the loss of sensitivity with dilution. The
manufacturer have data that dithiothreitol (DTT) significantly affects MSD readings at 10 mM
concentrations as it denatures antibodies, but this effect is minimal at 1 mM (Yvonne Clements,
personal communication). As sputum samples were at a final concentration of 5 mM DTE and we
assumed the effect of DTE would be similar to that of DTT. Conversely the manufacturer has
previously obtained no detectable cytokines when diluting DTE 5 fold. Spiking recovery was tested in
duplicate in an additional four samples at three concentrations: 10 pg/ml, 100 pg/ml and 1000 pg/ml of
cytokine standards.
Serum was not diluted beyond the standard protocol. The standard curve was made in Diluent 2.
Method (all incubation steps were at room temperature with vigorous shaking (125 rpm) throughout):
plates were incubated for 30 minutes with 25 μl of Diluent 2 per well, then 25 μL of sample diluted as
above or standards were added in duplicate and incubated for 2 hours. Plates were washed 3 times
Timothy SC Hinks 2. Materials and Methods
52
with 0.05% Tween 20 (Sigma) in PBS, incubated for a further 2 hours with 25 μl of detection antibody
and washed a further 3 times with PBS-Tween. 150 μl of 2x Read Buffer was added to each well and
the plate analysed immediately on a SECTOR Imager (MSD). Data were analysed in PRISM using a
4-parameter logistic model.
RNA extraction and quantitation
TRIzol
Ribonucleic acid (RNA) from MAIT clones was extracted using TRIzol® LS Reagent (Life
Technologies). 900-7000 cells were sorted directly into 1ml aliquots TRIzol and stored at -80 ̊C till
later use, at which point samples were defrosted and homogenized by repeated aspiration through a
1 ml filter-tip pipette, incubated at RT for 5 mins before addition of 200 μl of choloroform (Sigma).
After a further 5 minutes at RT samples were centrifuged at 16,200g, at 4 ̊C for 30 mins. The upper,
aqueous layer was transferred to a fresh tube containing 1 ml isopropanol (Fisher, Loughborough,
UK) and 5μg of glycogen and vortexed well, incubated at RT for 10 mins and centrifuged at 16,200g
at 4 ̊C for 30 mins to pellet the RNA. The supernatant was carefully removed and the pellet washed
with 1ml of 75% ethanol and incubated on ice for 10 mins, centrifuged at 16,200g at 4 ̊C for 5 mins
before all the ethanol was removed and the pellet dissolved in 30 μl diethylpyrocarbonate (DEPC)
treated water.
Nanoprep
For all other work RNA was extracted using the Absolutely RNA Nanoprep Kit (Stratagene). This kit
uses very small (10 μl volume) RNA-binding spin cups containing a silica-based matrix and is
optimised for purification of total RNA from very low cells numbers (1-104 cells).(Inc 2008) Our group
have previously used it successfully to perform PCR on as few as 10 sorted T cells(Vijayanand,
Seumois et al. 2007), and made minor modifications to the manufacturer’s protocol to maximise
sensitivity, specifically reloading samples onto matrix at several steps in the protocol and eluting into a
final volume of 15 μl rather than the recommended 10 μl (Vijayanand 2007).
Cells were flow-sorted directly into 100 μl aliquots of Agilent lysis buffer containing the chaotropic salt
guanidine thiocyanate to lyse cells and prevent degradation by RNases. Further protection from
RNases was provided by an additional 0.7 μl of 14.2M β-mercaptoethanol (β-ME) which reduces
disulfide bonds to irreversible denature RNases(Nelson 2005). Samples were vortexed hard for ≥10 s
then stored at -80 ̊C till further use.
Samples were defrosted and thoroughly mixed with an equal volume (100 μl) of 80% sulfolane
(Sigma), a water-soluble solvent and centrifuged in the spin-cup at 12,000g for 1 minute. Samples
were reloaded into the spin cup and centrifuged again at 12,000g for 1 minute and the filtrate
discarded. DNA was removed by digestion with DNase I in the following manner: spin cups were
washed with 300 μl of low-salt wash buffer, centrifuged once at 12,000g for 1 min, the filtrate
discarded and the spin-cup dried by centrifugation at 12,000g for 2 mins, then incubated for 15
Timothy SC Hinks 2. Materials and Methods
53
minutes at 37 ̊C with 15 μl of DNase I in digestion buffer solution. The RNA captured on the matrix
was then washed once with 300 μl high-salt wash buffer and twice with 300 μl low-salt wash buffer,
each time centrifuging at 12,000g for 1 min then discarding filtrate. Finally the RNA was eluted from
the matrix into a fresh collection tube by incubation for 2 minutes with 15 μl of elution buffer preheated
to 60 ̊C, with centrifugation at 12,000g for 2 minutes, followed by reloading of the eluate onto the
matrix and centrifuging for 12,000g for 5 minutes.
Nucleic acid quantitation
Nucleic acid concentration and purity were assessed with a NanoDrop 1000 spectrophotometer
(Thermo Scientific, Wilmington, USA). The concentration of nucleic acid was measured in 1 μl of
sample and the purity assessed by determination of the ratio of sample absorbance at 260 and 280
nm: pure DNA has a ratio of 1.8 and pure RNA a ratio of 2.0, although the 260/280 ratio is also
affected by changes in pH and differing nucleotide mixes in the nucleic acid(ThermoScientific 2008).
Reverse transcription and polymerase chain reaction
Reverse transcription with SuperScriptTM III RT kit
RNA from MAIT clones was reverse transcribed (RT) using the SuperScriptTM III reverse transcriptase
kit (18080-093, Invitrogen) which uses a modified pol gene of Moloney Murine Leukemia virus, in a
reaction volume of 20 μl. 11 μl of RNA were incubated with 1 μl random primers (predominantly
hexamers) and 1 μl deoxyribonucleotide triphosphate (dNTP) mixture at 65 ̊C for 5 minutes followed
by immediate quenching on ice to remove secondary structure of the RNA. A no template control
contained 11 μl of ddH20 in place of RNA. To this was added 4 μl of 5x first strand buffer (250 mM
Tris-HCl (pH 8.3 at RT, 375 mM KCl, 15 mM MgCl2), 1 μl 0.1M DTT, 1 μl RNase inhibitor (RNase
OUTTM) and 1 μl of the RT enzyme. Samples were mixed by pipetting, incubated at RT for 5 mins,
then incubated in a Tetrad 2 DNA engine(MJ Research, Bio-Rad, Hemel Hempstead, UK), at 50 ̊C for
60 mins, followed by enzyme inactivation at 70 ̊C for 15 mins.
Reverse transcription with Precision nanoScriptTM RT kit
RNA from sorted airway T cells was reverse transcribed using Precision nanoScriptTM reverse
transcription kit (formerly known as qScript, Primer Design, Rownhams, UK), which also uses the
same RT enzyme, but for priming used a mixture of both oligo-dT priming and random nonomers.
Oligo-dT primers bind to the polyA tail of messenger RNA (mRNA), preferentially targeting the 3’end
of mRNA and so reducing transcription of ribosomal RNA (rRNA), which is advantageous for
quantitative PCR (qPCR) as it leads to lower threshold cycle (CT) values. The advantage of
incorporating random nonomers is increased priming efficiency with partially degraded RNA, which
was important to achieve maximum sensitivity for my work on formalin fixed cells(PrimerDesignLtd).
12.5 μl of RNA were incubated at 65 ̊C for 5 minutes followed by 1 minute on ice. A no template
control contained 12.5 μl of ddH20. To this was added 0.5 μl random nonomers, 1 μl oligoDT, 1 μl
dNTP mixture, 2 μl of 100 mM DTT, 2 μl of 10x buffer and ddH20 to a volume of 20 μl. 1 μl of the RT
Timothy SC Hinks 2. Materials and Methods
54
enzyme was added to each sample, or 1 μl ddH20 for the “no enzyme control” to test for genomic
contamination. Samples were mixed and incubated in a Mastercycler (Eppendorf, Hamburg,
Germany) at 55 ̊C for 20 mins, then 75 ̊C for 15 mins.
Polymerase chain reaction (PCR)
For non-quantitative work on MAIT cell clones, RT products were amplified by polymerase chain
reaction (PCR) using BioTaq DNA Polymerase (BIO-21040, Gentaur, Brussels, Belgium). The
reaction mixture, 50 μL total, contained 1 μl of cDNA, 5 μl 10x NH4 buffer, 2.5 μl of 50 mM MgCl2, 1μl
of 10mM dNTP, 0.5 μl of BioTaq DNA polymerase, 1 μl of forward primer, 1 μl of reverse primer (final
concentration of 2 μM in reaction) and 39 μl of ddH20.
Amplifications were performed on the Mastercycler with the following cycling parameters: 96 ̊C for 1
min preincubation, then 96 ̊C for 30 s denaturing, then 65 ̊C for 30 s annealing, then 72 ̊C for 2
minutes elongation; repeated for 45 cycles and finished with a 10 min extension at 72 ̊C.
Primers (MWG Biotech AG, Ebersberg, Germany) are shown in table 2.3 and were designed to span
the canonical Vα7.2-Jα33 rearrangement of the CDR3α region of the TCRα chain(Porcelli, Yockey et
al. 1993; Tilloy, Treiner et al. 1999).
Table 2.3 Oligonucleotide primers used for PCR
Product Sequence
Vα7.2 TCR 5’-ATA TAT CAT ATG GGA CAA AAC ATT GAC CAG-3’ fwd
Jα33 TCR 5’-GCT TTA TAA TTA GCT TGG TCC CAG C-3’ rev
IL17-A 5’-CCT CAG ATT ACT ACA ACC GAT CC-3’ fwd
5’-CAC TTT GCC TCC CAG ATC AC-3’ rev
FOXP3 5’- CAG CAC ATT CCC AGA GTT CCT-3’ fwd
5’- GCG TGT GAA CCA GTG GTA GAT-3’ rev
β-2 microglobulin* Accession number: NM_004048.2
Anchor nucleotide: 362
Context sequence length: 141bp
YWHAZ* Accession number: NM_003406.3
Anchor nucleotide: 2585
Context sequence length: 150bp
*The exact sequences are commercially sensitive and not released by the manufacturer.
Timothy SC Hinks 2. Materials and Methods
55
Gel electrophoresis
10 μL of PCR product premixed with dye was loaded into wells of a 1% (w/v) agarose (Melford,
Chelsworth, UK) gel in Tris/Borate/EDTA (TBE) buffer containing 10 μl per 100 ml of Nancy-520 dye
(01949, Sigma) and electrophoresed at 80 volts for 60 minutes, before photographing the gel under
ultraviolet light.
Quantitative PCR
Quantitative PCR (“qPCR” or “real-time PCR”) was used to measure the abundance of mRNA
transcripts for IL17-A and FOXP3, as well as the normalising genes B2M and YWHAZ. The reaction
mixtures, performed in duplicate 20 μl volumes contained 9 μl of cDNA (or ddH20 for “no template”
control), 10 μl of PrecisionTM Mastermix (Primer Design, containing a thermostable Taq polymerase)
and 1 μl of forward and reverse primer mix (6 pmol of each). qPCR analysis was performed using the
iCyclerIQ platform (Bio-Rad) with the following cycling parameters: 95 ̊C for 10 min preincubation,
then 95 ̊C for 10 s, then 50 ̊C for 30 s, then 72 ̊C for 10 seconds; repeated for 52 cycles.
PerfectProbeTM primer pairs were designed and tested for amplification efficiency by Primer Design.
They are hydrolysis probes in which a quencher molecule at the 3’ end of the probe reduces the
fluorescence of a fluorophore (FAM-490) at the 5’ end of the molecule via fluorescence resonance
energy transfer(Holland, Abramson et al. 1991). PerfectProbe differ from this original description in
that the quencher and fluorophore are brought into closer proximity by a hairpin loop structure,
providing more efficient quenching thus a lower background (PrimerDesignLtd).
Flow cytometry
Surface staining for MAIT cells
Cells were transferred to polypropylene test tubes in + 1 ml of PBS, centrifuged at 400g for 5 mins 4 ̊C
and resuspended in 500 μl PBS on ice. Cells were stained with 1 μl of LIVE/DEAD® Fixable Violet
Dead Cell Stain for 405 nm excitation (L34955, Invitrogen) for 30 mins, then washed with 2 ml
magnetic activated cell sorting (MACS) buffer, centrifuged at 400g for 5 mins and resuspended in 200
μl of MACS buffer for surface staining for 30 mins. Cells were then washed with 2 ml of MACS buffer,
centrifuged at 400g for 5 mins and resuspended in 200 μl of MACS buffer for cytometry the same day.
All staining protocols were performed in the dark and on ice, with a centrifuge refrigerated to 4 ̊C.
Timothy SC Hinks 2. Materials and Methods
56
Table 2.4 Antibodies and fluorochromes used for surface staining
Stain Clone Supplier,
reference
Ex-MAX / Em-
MAX
wavelength (nm)
Concentration (μl)
PBMC Other
tissues
LIVE/DEAD® Fixable
Violet Dead Cell Stain
N/A Invitrogen,
L34955
405/450 1 1
CD3 PE-CyTM7 SK7 BD, 557851 496,546/785 1.5 3
CD4 PerCP-CyTM5.5 L200 BD, 552838 482/695 2.5 5
CD8 APC-CyTM7 SK1 BD, 348813 650/785 2.5 5
TCR Vα7.2 PE 3C10 Biolegend,
351706
496, 546/578 5 10
CD161 FITC DX12 BD, 556080 494/519 10 10
γδTCR FITC B1 BD, 61995 494/519 5 5
Table 2.5 Isotype controls
Fluoroch
rome
Stain for which
this is used as
control
Isotype
class
Clone Supplier,
reference
Concentration (μl)
PBMC Other
tissues
FITC CD161 or
γδTCR
Mouse IgG1κ MOPC-21 BD, 555748 10 10
FITC IL-13 Mouse IgG1 11711 R&D Systems,
IC002F
10 10
PE TCR Vα7.2 Mouse IgG1κ MOPC-21 Biolegend,
400114
0.25 0.5
PE IL17 Mouse IgG1 P3.6.2.8.1 eBioscience,
12-4714-42
2.5 2.5
APC IFN-γ or TNFα Mouse IgG1κ P3.6.2.8.1 eBioscience
17-4714-42
0.3 0.3
APC FOXP3 Rat IgG2α κ eBR2a eBioscience,
17-4321
2.5 2.5
Rat
serum
N/A To block nonspecific
staining by FOXP3 APC
eBioscience,
245555
2 2
Timothy SC Hinks 2. Materials and Methods
57
Intracellular cytokine staining
Cells at a concentration of 1x106 cells/ml were first stimulated at 37 ̊C, 5% CO2 for 4-5 hours (see
table 2.6) with 25ng/ml phorbol 12-myristate 13-acetate (PMA, Sigma) which activates protein kinase
C and 500ng/ml ionomycin (Sigma) a Ca2+ ionophore, in the presence of 2μM monensin
(eBioscience), an inhibitor of trans-Golgi function(Nylander and Kalies 1999), leading to intracellular
accumulation of cytokines. Paired samples were left unstimulated, without PMA, ionomycin or
monensin, both to act as negative controls for the intracellular cytokine staining and also for FOXP3
analysis. Biopsy tissue was stimulated as complete biopsies in 0.5 ml wells of 24-well plates but after
4 hours collagenase dispersed for 1 hour at 37 ̊C, before undergoing the same staining as other
tissues.
Table 2.6 Stimulation times for each tissue
Tissue Duration of ex vivo stimulation
Fresh PBMC 5 hours
Sputum 4.5 hours
Bronchoalveolar lavage 5 hours
Bronchial biopsies 4 hours + 1 hour collagenase digestion
Previously cryopreserved PBMC 5 hours
Next cells were resuspended and transferred to polycarbonate test tubes in + 1ml of PBS, centrifuged
at 400g for 5 mins 4 ̊C and resuspended in 500 μl PBS on ice. Cells were stained with 1 μl of
LIVE/DEAD® Fixable Violet for 30 mins, then washed with 2ml MACS buffer and centrifuged at 400g
for 5 mins. Cells were then fix-permeabilised by resuspending in 200 μl of fixation/permeabilization
working solution (eBioscience, comprising 1 part Fixation/Permeabilization Concentrate 00-5123 and
3 parts Fixation/Permeabilization Diluent 00-5223) for 30 mins, washed with 2 ml of permeabilisation
buffer (00-8333, eBioscience, diluted 1:9 with MACS buffer) and centrifuged at 400g for 5 mins. Cells
were then resuspended in residual volume, adjusted to total 110 μl volume diluted permeabilisation
buffer and incubated with fluorochrome-conjugated antibodies for 45 mins on ice in the dark, before a
final wash with 2 ml permeabilisation buffer, centrifugation at 400g for 5 mins and resuspension at
≥200 μl volume for flow cytometry.
Timothy SC Hinks 2. Materials and Methods
58
Table 2.6.1 Antibodies and fluorochromes used for intracellular staining
Stain Clone Supplier,
reference
Ex-MAX / Em-
MAX
wavelength
(nm)
Concentration (μl)
PBMC Other
tissues
IL17A PE eBio64CAP17 eBioscience,
12-7178-42
496, 546/578 5 5
IL13 FITC 32007 R&D Systems,
IC2131F
494/519 5 5
IFN-γ APC 4S.B3 eBioscience,
17-7319-82
650/660 0.3 0.3
FOXP3 APC PCH101 eBioscience,
17-4776-42
650/660 5 5
TNFα APC MAb11 eBioscience,
17-7349
650/660 0.5 0.5
Cell sorting and data acquisition
Flow cytometry was performed with a nine-colour FACS AriaTM cell sorter (BD Biosciences) with three
lasers at 488, 633 and 407nm wavelengths. Samples other than PBMC were passed again through a
70 μm prior to acquisition on the flow cytometer. Purity was checked using FACS Accudrop beads
(BD) and was ≥99% in all cases. Cells were sorted, at 50 μl/min, directly into eppendorfs containing
lysis buffer, except PBMC T cells where the sample volume required an additional centrifugation step
at 400g for 5 mins to pellet cells and remove the supernatant prior ot addition of lysis buffer. Samples
were acquired, sorted and analysed using FACS DivaTM 5.0.3 software (BD) and the following conflict
resolution settings: yield mask 0, purity mask 32 and phase mask 0.
Gating strategy for MAIT cells
T cells were identified by pulse width-pulse area doublet exclusion, dead cell exclusion by Violet
LIVE/DEAD and by side-scatter and CD3-PE-Cy7 staining. MAIT cells were defined as cells double
positive for CD161-FITC and the TCR Vα 7.2 chain conjugated to PE (Dusseaux, Martin et al. 2011).
Gates were set using isotype controls for IgG1κ-FITC and IgG1-PE (Figure 2.6).
Timothy SC Hinks 2. Materials and Methods
59
Figure 2.6 Gating strategy for MAIT cells
T cells were identified by pulse width-pulse area doublet exclusion, dead cell exclusion by Violet
LIVE/DEAD and by side-scatter and CD3-PE-Cy7 staining. MAIT cells were defined as cells double
positive for CD161-FITC and the TCR Vα 7.2 chain conjugated to PE. Gates were set using isotype
controls for IgG1κ-FITC and IgG1-PE.
Timothy SC Hinks 2. Materials and Methods
60
Gating Strategy for T helper cells
Doublets were excluded by gating on forward scatter-area (pulse area) (Figure 2.7, A), which is
proportional to the cross sectional area of the cell, versus forward scatter-width (pulse-width), which is
proportional to the time taken to pass through the laser beam. Pulse width is greater in doublets,
which become aligned perpendicularly to the laser beam in a narrow stream. Dead cells were
excluded by their increased uptake of Violet LIVE/DEAD® Fixable (Figure 2.7, B), which permeates
compromised membranes of necrotic cells to react irreversibly with free amines intracellularly
(Invitrogen 2007). T cells were identified according to surface staining by CD3-PE-Cy7 and side
scatter used to exclude large cells and debris (Figure 2.7, C). T cells were further characterised by
their surface staining with CD4-PerCP-Cy5.5 and CD8-APC-Cy7 (Figure 2.7, D) and intracellular
staining for IL-17PE (TH17 cells), IFN-γ APC (TH1 cells) and IL-13 FITC (TH2 cells)(Figure 2.7, E).
Gates for each of these cytokines were set on unstimulated cells using the same stains. Regulatory T
cells were identified by intracellular staining for FOXP3 APC.
Controls for flow cytometry
As there is a 50-fold difference in fluorescence between live and dead cells stained with Violet
LIVE/DEAD fixable (Invitrogen 2007), two clear fluorescence peaks were always distinguishable to
enable confident setting of the dead cell gate, reinforced by the use of an unstained control (Figure
2.8, A). As CD3 is abundantly expressed on the surface of T cells, staining was universally bright and
unstained cells could be used as negative controls for the CD3+ gate (Figure 2.8, B).
Timothy SC Hinks 2. Materials and Methods
61
Figure 2.7 Gating strategy T helper cells
Schematic to demonstrate the hierarchical gating strategy employed for flow-cytometric analysis of
CD4+ T cells throughout the study. Doublets were excluded by gating on forward scatter-area (pulse
area) versus forward scatter-width (pulse-width)(A). Dead cells were excluded by their increased
uptake of Violet LIVE/DEAD® Fixable (B) T cells were identified according to side scatter profile and
surface staining by CD3-PE-Cy7 (C). T cells were characterised by CD4-PerCP-Cy5.5 and CD8-APC-
Cy7 (D), and intracellular staining for IL-17PE, IFNγ APC and IL-13 FITC (E). Gates for each of these
cytokines were set on unstimulated cells using the same stains. TREG were identified by intracellular
staining for FoxP3 APC.
Timothy SC Hinks 2. Materials and Methods
62
Figure 2.8 Controls for cytometry
With Violet LIVE/DEAD fixable two clear fluorescence peaks were always distinguishable, reinforced
by the use of an unstained control (A). CD3-PE-Cy7staining was universally bright with unstained
cells as negative controls (B). Gates for TH17, TH1 and TH2 cells were set using unstimulated cells
(rested overnight in culture in absence of PMA, ionomycin or monensin), although initially isotype
controls were also included (C, D), but could provide spurious results (D middle panel).
Timothy SC Hinks 2. Materials and Methods
63
Setting of gates for intracellular stains is more challenging as there are rarely clearly demarcated
populations within a single specimen. Three main approaches include use of isotype controls,
“fluorescence minus one” technique, or unstimulated controls stained with the same antibodies. A
number of well recognised problems in the use of isotype controls mean they are frequently of little
value, since each antibody and antibody conjugate has very different background staining
characteristics (Baumgarth and Roederer 2000). Specifically isotype controls do not have identical
tertiary structure, there may be subtle differences because they originate from hybridomas which are
not normal cells, the binding of isotypes varies between cell types and their stage of differentiation
and variations in purification methods or efficiency of conjugation cause significant lot to lot variation
(Keeney, Gratama et al. 1998). For these reasons use of isotypes is particularly problematic for rare
events (Keeney, Gratama et al. 1998). These problems can be avoided by the alternative technique of
fluorescence minus one the sample is divided into multiple aliquots and each stained with all reagents
except for the one of interest (Baumgarth and Roederer 2000; Roederer 2001), but this would not be
practicable for this study due to the very small sizes of the tissue samples and the large number of
stains. The third approach is to divide a single sample into two equal parts during overnight culture,
one of which receives no stimulation with PMA, ionomycin and no Golgi block, preventing any
intracellular accumulation of cytokine, but was then stained with an identical panel of stains to the
stimulated control. In practice I used this technique which consistently provided reliable negative
populations on which to set gates (Figure 2.8, C and D). Initially I also used isotype controls in each
case (Figure 2.8, C), but eventually abandoned this as it wasted samples and proved less reliable,
frequently providing inaccurate or frankly spurious results, such as higher staining than that seen with
the specific antibody (Figure 2.8, D, Figure 2.9).
Timothy SC Hinks 2. Materials and Methods
64
Figure 2.9 Comparison of isotypes and unstimulated cells
A schematic showing additional comparison of the use of isotypes or unstimulated cells as negative
controls for determining gating on intracellular cytokine stains. For rare populations such as TH2 cells
in sputum it is frequently more accurate to gate on unstimulated cells than using isotype controls,
which here would give a negative value for the TH2 cell frequency (A). TH17 and TH1 cells are more
abundant and with brighter staining, so isotypes could have been an acceptable alternative (B). As it
is theoretically possible that apparent specific IL-17-PE staining might have been an artefact of
increased non-specific binding due to the effect of stimulation, additional positive and negative
controls are shown in (C). Stimulated cells were stained for IL-17 according to the usual protocol
either in the presence of a saturating quantity of recombinant human IL-17 (rhIL-17) to block all
specific binding of the antibody (negative control), or after pre-incubation of the cells with rhIL-17 to
Timothy SC Hinks 2. Materials and Methods
65
cause saturation of the cells’ IL-17 receptors (positive control). Together these demonstrate that
events with high PE fluorescence do indeed represent true staining for cells expressing IL-17.
Cloning of MAIT cells
As part of my analysis of MAIT cells I sought to clone these cells. This was done with help from a
post-doctoral fellow, Dr Salah Mansour, working with Professor Gadola. The method we used was
adapted from a protocol developed for cloning iNKT cells (Matulis, Sanderson et al. 2010). PBMC
were isolated from a healthy individual and stained with antibodies to detect live CD3+CD161+Vα7.2+
cells (MAIT cells) which were sorted at 1 cell/well into a 96-well round-bottom plate containing 5x104
autologous irradiated (30 Gy) PBMC feeder cells in 200 μl of T cell sorting medium (see above,
contains 10% foetal calf serum, 2 mM L-glutamine, 2 mM sodium pyruvate, 100 µg/ml streptomycin,
100 U/ml penicillin, MEM essential amino acids, non-essential amino acids, 400 U/ml rh-IL2, 10 ng/ml
rhIL-7, 10 ng/ml rhIL-15, 2% human serum, 0.1 mg/ml Kanamycin, 50 μM β-ME) and also
supplemented with PHA (Sigma) at 1 μg/ml and then cultured at 37 ̊C 5% CO2 for several weeks.
After 2 weeks culture expanded clones were transferred sequentially into 24-well, 12-well, and 6-well
flat-bottom plates in fresh T cell growth medium. Culture medium was refreshed every 2-5 days when
a colour change was observed, and at day 29 clones were restimulated with PHA 1 μg/ml. Expanded
clones were tested by flow cytometry for surface phenotype and by PCR to confirm the presence of
the canonical Vα7.2-Jα33 TCR rearrangement.
Definition of T helper cells for flow cytometry
Accurate enumeration of T cell subsets by flow cytometry depends on accurate definitions of these
populations according to consistent and logical setting the gates used in the hierarchical Boolean
gating analysis. For the work in my thesis I had to address some specific challenges particular to
individual cell populations. Firstly accurately defining CD4+ T cells in the light of stimulation induced
downregulation of the CD4 co-receptor and secondly determining the exact definition of a positive
FOXP3 gate necessary to define regulatory T cells. I addressed these in turn in the following
optimisation experiments.
The problem of CD4 co-receptor downregulation during ex-vivo stimulation
T helper lymphocytes are defined by their expression of the CD4 co-receptor (Reinherz, Kung et al.
1980; Reinherz and Schlossman 1980; Bernard, Gay-Bellile et al. 1984). However it is well
recognised that stimulation of cells, such as with PMA/ionomycin, causes downregulation of surface
receptors, including CD3, CD8 and particularly CD4(O'Neil-Andersen and Lawrence 2002; Hawn,
Misch et al. 2007). As the objective of my work was to analyse CD4+ and CD8+ T cell populations
using intracellular cytokine staining which depends on ex vivo cell stimulation, this effect could have
proved problematic. I therefore analysed the effects of stimulation on CD4+ and CD8+ T cell
subpopulations in different cellular compartments to address the question of whether it is preferable to
define T helper cells by positive selection on CD4 or negative selection on CD8.
Timothy SC Hinks 2. Materials and Methods
66
Results
Figure 2.10 shows changes in each of the 4 subpopulations of T cells defined by CD4 and CD8
expression before and after stimulation for 4-5 hours with PMA/ionomycin. Changes in the proportion
of T cells positive for CD4 are modest, with no significant change in BAL, a significant, but modest
decrease in sputum CD4+ cells from 65% (60-74, median and IQR) unstimulated to 46% (37-55) after
stimulation (P=0.0005). This is contrasted with a slight increase in CD4+ cells in PBMC from 49% (44-
52) to 59% (44-61, P=0.008, see table 2.7), which might perhaps be due to preferential death of
CD4+ cells from monensin toxicity(Nylander and Kalies 1999).
Timothy SC Hinks 2. Materials and Methods
67
Figure 2.10 Changes in CD4+ and CD8+ populations with stimulation
Changes in each of the 4 subpopulations of T cells defined by CD4 and CD8 expression before and
after stimulation for 4-5 hours with PMA/ionomycin. n=10 for each compartment, significance tested
with paired t tests, not corrected for multiple comparisons.
Timothy SC Hinks 2. Materials and Methods
68
The phenomenon of CD4 downregulation is variable between subjects. Whilst it increases over time,
being more marked at 5 than 4 hours, Figure 2.11 A,B shows how samples from two individuals differ
markedly in the extent of CD4 internalisation. To what degree does CD4 downregulation adversely
affect the purity of my T helper cell population if we positively select on CD4? I compared the
percentage of cells within the CD4+ gate which were also CD8+ before and after stimulation and
found these frequencies were low and did not change significantly with stimulation. The proportions of
CD4+ cells also expressing CD8 were 0.75% (0.40-1.7) unstimulated and 0.50% (0.13-1.6) after
stimulation in PBMC (p=0.2) and in BAL were 0.65% (0.23-1.4) unstimulated and 0.85% (0.53-2.8)
stimulated (P=0.1) and in sputum were 3.6% (1.0-4.5) unstimulated and 2.9 (2.7-6.2) (P=0.2). See
Table 2.7.
Table 2.7 The effect of stimulation on relative T cell populations in different tissues.
PBMC BAL SputumUnstimulated Stimulated P Unstimulated Stimulated P Unstimulated Stimulated P
0.75 (0.40‐1.7) 0.50 (0.13‐1.6) 0.2 0.65 (0.23‐1.4) 0.85 (0.53‐2.8) 0.1 3.6 (1.0‐4.5) 2.9 (2.7‐6.2) 0.2
CD4+ (%) 49 (44‐52) 59 (44‐61) 0.008 62 (41‐76) 41 (37‐51) 0.001 65 (60‐74) 46 (37‐55) 0.0005CD8+ (%) 34 (28‐44) 28 (15‐36) 0.0005 29 (20‐36) 28 (17‐36) 0.1 23 (17‐24) 18 (14‐26) 0.2CD4+8+(%) 0.4 (0.33‐1.2) 0.4 (0.20‐0.75) 0.4 0.4 (0.33‐0.58) 0.55 (0.40‐1.1) 0.2 2.2 (0.73‐3.7) 2.3 (0.70‐3.9) 0.5
CD4‐8‐ (%) 18 (13‐24) 18 (13‐25) 0.9 7.8 (6.0‐11) 27 (20‐37) 0.0002 8.8 (3.63‐17) 27 (20‐43) 0.0002
CD4/8 ratio 1.4 (1.0‐2.0) 2.1 (1.2‐4.5) 0.004 2.2 (1.1‐3.7) 1.4 (1.1‐2.9) 0.009 2.9 (2.2‐4.4) 2.8 (1.5‐3.1) 0.07
Table shows median and IQRs. Uncorrected P values are for paired T tests, with significant values in bold.
Purity of CD4 population (%of CD4 +cells which are CD8+)
Timothy SC Hinks 2. Materials and Methods
69
Figure 2.11 CD4 receptor down regulation
CD 4 co-receptor expression before and after 4.5 hours stimulation with PMA/ ionomycin. Despite
poorer discrimination of the CD4+ and CD- populations, purity from CD8+ cells is usually
preserved.(A) An extreme case of CD4 down-regulation after 5 hours stimulation with PMA/
ionomycin (B). Despite this dramatic change, the CD4 gate retains high purity from CD8- cells.
Moreover it can be shown that the majority (71.3% in this case) of IL17+ cells continue to fall within
the CD4+ gate (C). A more generally representative plot of CD4 expression amongst IL17+ cells
showing that the majority of IL17+ fall within the CD4+ gate: nearly 90% in this case. Thus those cells
which move out of the CD4+ gate due to receptor down-regulation are rarely IL17+ cells of interest.
Defining the T helper cell population by negative selection on CD8 would have the benefit of being
free from CD4 downregulation. However, it would instead be affected by CD8 downregulation, albeit
Timothy SC Hinks 2. Materials and Methods
70
to a lesser extent as the receptor is subject to less change of expression. More importantly it would
necessarily include the population of CD4-8- cells which comprise 0.4-2.3% of all T cells. This would
outweigh the benefit of avoiding use of CD4 as a marker. Furthermore it can be shown that whilst
some cells disappear from the CD4 subset during stimulation, these tend not to be those which are
strong cytokine producers. See Figure 2.11C. Specifically, because the brightest CD4 cells produce
the strongest cytokine secretion, the vast majority of TH17 cells continue to fall within the CD4+ gate
despite CD4 downregulation: nearly 90% in the case shown and in excess of 70% of cells fall within
this CD4+ gate even in the most extreme cases of receptor downregulation. See Figure 2.11 B,C.
Conclusion
Positively selecting on CD4+ cells in PBMC, BAL and sputum avoids contamination with double
negative T cells, was affected only modestly by CD4 downregulation and did not increase
contamination by CD8+ cells. I therefore used this strategy, except in the case of bronchial biopsies,
which were additionally affected by collagenase cleavage of CD4.
Definitions of Treg for flow cytometry
Our current concept of regulatory T cells is based on Sakaguchi’s 1995 description murine
CD4+CD25+ regulatory T cells (Sakaguchi, Sakaguchi et al. 1995). Baecher-Allan later demonstrated
that the human equivalent of these murine CD4+25+ cells was the CD4+CD25Hi population (Baecher-
Allan, Brown et al. 2001), i.e. a population expressing high levels of CD25 that have the capacity to
suppress the function of effector T cells. In 2003 Hori et al described the master regulator
transcription factor FOXP3 as governing the development and function of Tregs and it is now
considered the most reliable marker for natural Tregs (Hori, Nomura et al. 2003). Deficiencies in
FOXP3 cause the immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX
syndrome) (Hori, Nomura et al. 2003), with autoimmune responses in multiple organs in both humans
and mice due to loss of peripheral tolerance (Ostroukhova, Qi et al. 2006). Expression of the FoxP3
gene in transgenic mice and ectopic expression of FOXP3 in human cells can genetically reprogram T
cells to a regulatory phenotype (Fontenot, Gavin et al. 2003; Hori, Nomura et al. 2003; Khattri, Cox et
al. 2003).
Problems with existing markers
CD25
As CD25 is a surface marker, it has proved to be a useful. However, because CD25 is also a marker
of activation of T cells, its specificity as a marker of Tregs is limited. (Baecher-Allan, Brown et al.
2001). Furthermore, whilst the CD25hi population is more specific for natural Tregs, it lacks sensitivity
because a significant proportion of FOXP3 expressing cells are CD25-. Whilst only 1-2% of peripheral
CD4+ cells are CD25hi, up to 8-10% of peripheral CD4+ T cells in humans are FOXP3 positive (Liu,
Putnam et al. 2006). Thus the CD4+25hi definition lacks sensitivity and specificity.
Timothy SC Hinks 2. Materials and Methods
71
CTLA4 and GITR are also reported to be expressed on Treg, but as with CD25 are also expressed on
effector T cells (Liu, Putnam et al. 2006).
CD127
In 2006 two groups described CD4+25+CD127lo cells in humans as expressing high levels of FOXP3
and having suppressive functions(Liu, Putnam et al. 2006; Seddiki, Santner-Nanan et al. 2006).
These cells are anergic and as suppressive as CD25Hi cells, but are three times as abundant as
CD25Hi cells (Liu, Putnam et al. 2006). Liu et al also showed an inverse correlation between FOXP3
and CD127 expression and showed FOXP3 interacts with the CC127 promoter.
In healthy adults Seddiki et al found CD4+127lo was 87% specific and 84% sensitive for
CD25+FoxP3+ cells and that suppression was cell-contact dependent (Seddiki, Santner-Nanan et al.
2006).
In summary, FOXP3 remains the most specific marker for natural Treg (Liu, Putnam et al. 2006).
Initially relative FOXP3 mRNA expression was measured (Guyot-Revol, Innes et al. 2005), but more
now antibodies are available, allowing me to use intracellular staining. Nonetheless FOXP3 is only an
intracellular marker, so this technique requires fixation and permeabilisation preventing isolation of
viable Tregs that are needed in functional studies. For these studies CD127lo is superior to CD25hi
cells, but the best surface phenotype is probably CD4+25+127Lo, which comprise 87% of FOXP3+
cells (Seddiki, Santner-Nanan et al. 2006).
Frequencies of Treg
CD4+CD25+ Treg cells comprise approximately 10% of peripheral CD4+ cells(Roncador, Brown et al.
2005). Expression of FOXP3 is highly restricted to Treg populations - indeed genetic transfer of
FoxP3 converts naive CD4+CD25- T cells to a regulatory phenotype – and is highly correlated with
CD25 expression in natural CD25+ Treg, so it is a useful specific marker of Treg cells(Wang, Zhang
et al. ; Roncador, Brown et al. 2005). Using the phenotype CD3+CD4+FOXP3+ to define Treg the
upper range of Treg in peripheral blood is 8-10% of CD4+ T cells in humans (Liu, Putnam et al. 2006;
Lin, Chen et al. 2007; Bonelli, von Dalwigk et al. 2008; Bi, Suzuki et al. 2009). The normal range in
health has been estimated at 7.5%±2.4% (mean ±SD) (Bi, Suzuki et al. 2009) or 6.5%±1.3%(Bonelli,
von Dalwigk et al. 2008). This equates to approximately 4% of CD3+ cells (range 2.5-7.5) (Brusko,
Wasserfall et al. 2007)or 1.2% of all lymphocytes(Freier, Weber et al.).
For my work I defined Treg as live, singlet, CD3+, CD4+, FOXP3+ cells. It was not practical to use
CD25 as an additional marker due to the small size of tissue samples and use of other cytometer
channels for markers for other T cell subsets. Isotype controls for rat IgG2α-APC proved less reliable
for setting the FOXP3+ gate than use of the CD4 negative population, which does not express
FOXP3, to act as a negative control within the same sample tube (see Figure 2.11). This approach
was based on that previously used by my colleague Asha Ganesan (Ganesan 2010). To ensure
Timothy SC Hinks 2. Materials and Methods
72
consistency it was necessary to define the number of non-specific events which could be found within
this empty P1 gate as a proportion of total CD3+ cells. Due to cleavage of CD4 by collagenase,
biopsy Treg were defined as the percentage of live FOXP3+ CD8- T cells, using an empty P2 gate set
on CD8+ cells.
Figure 2.12 Setting of regulatory T cell gates
Schematic of hierarchical gating strategy used to identify Treg, which were defined as percentage of
live singlet, CD3+CD4+ cells expressing FOXP3. As FOXP3 was not expressed in CD4- cells this
gate was set by the level amongst CD4- cells, such that the P1 gate contained ≤0.5% of CD3+ cells.
Biopsy Treg were identified in an analogous manner as percentage of live, singlet CD3+8- T cells
expressing FOXP3, according to a CD8+P2 gate containing ≤0.5% of CD3+ cells.
Timothy SC Hinks 2. Materials and Methods
73
Data analysis to define set-point of FOXP3+ gate
To define this set-point I analysed data I acquired from PBMC from 15 healthy individuals. I analysed
T cell frequencies using cut-points of ≤0.5, ≤0.4, ≤0.3, ≤0.2.
Results
Results are shown in Table 2.8
Table 2.8 Results of Treg set-point analysis. PBMC T cell frequencies in n=15 healthy
individuals according to definition of P1 gate.
Treg frequency
(% of total CD4 T cells)
P1 cut point (% of total CD3+ cells within gate)
≤0.5 ≤0.4 ≤0.3 ≤0.2
Median 4.5 4.1 3.7 3.4
Interquartile range (3.8-7.5) (3.4-6.4) (3.1-5.6) (2.4-5.0)
Maximum and minimum range (1.3-11) (1.1-11.2) (0.9-10.5) (0.6-9.9)
Conclusion
These results imply that adjusting the cut-point has rather little effect on the maximum and minimum
values, implying that data-points occurring at the extremes of the Treg range were true values, rather
than artefacts of poorly chosen cut-points. This conclusion is supported by the wide ranges seen in
the studies cited above.
A cut point of P1 ≤0.5% would be consistent with previous work from our group (Ganesan 2010) and
gives a mean (±SD) of 5.2%±2.7% which would be consistent with, but on the low side of the normal
ranges for the studies cited above. Therefore I selected a cut-point of up to ≤0.5% in P1 gate for
PBMC, sputum and BAL and of ≤0.5% in the P2 gate for biopsy T cells.
Comparison of fresh versus cryopreserved PBMC
The need for cryopreservation
My analysis of viral induced exacerbations was based on samples obtained from the SG005 clinical
study already described, in which samples were obtained on a daily basis for a period of 13 months
and which therefore made it necessary to cryopreserve peripheral blood mononuclear cells to
minimise inter-assay and inter-operator variability. It was however necessary to optimise the methods
used for cryopreservation and for subsequent intracellular cytokine staining and also to determine
what effect this additional step had on my immunological results. What follows is the optimisation and
validation work I conducted.
Cryopreservation has become a widespread technique in recent years (Disis, dela Rosa et al. 2006)
and is particularly valuable in longitudinal studies of immunological parameters as it minimises inter-
Timothy SC Hinks 2. Materials and Methods
74
assay variability inherent in studies, where assays will be affected by changes in unstable reagents,
or variability in batches of reagents and by variability in the conduct of an experimental
protocol(Weinberg, Song et al. 2009). Secondly it minimises inter-operator variability, as it can enable
all assays to be performed by a single investigator, despite sample collection occurring on a daily
basis over a long period, by a number of users. Thirdly, the ability to batch process samples greatly
reduces workload, making practicable a volume of work which would otherwise be unviable.
Concerns regarding cryopreservation
The method of cryopreservation can have a significant impact on cell viability and functional
responses (Betensky, Connick et al. 2000; Weinberg, Wohl et al. 2000; Maecker, Moon et al. 2005),
and may increase CD4+ T cell apoptosis (Owen, Sinclair et al. 2007). Moreover, the technique itself
may introduce a new source of inter-assay variability. It was thus essential to ensure the technique
was optimal prior to collecting longitudinal samples. Nonetheless, with optimal cryopreservation
techniques there is evidence of good correlation between results obtained from intracellular cytokine
staining with fresh cells and those obtained using cryopreserved cells, with results which did not differ
significantly (Maecker, Moon et al. 2005; Jeurink, Vissers et al. 2008; Weinberg, Song et al. 2009).
Similarly several groups have demonstrated strong correlation between Treg frequencies in fresh and
frozen PBMC (Costantini, Mancini et al. 2003; Elkord 2009).
Method
PBMC were obtained from healthy individuals and either resuspended in RPMI with 10% FCS for
overnight culture at 37 ̊C, 5%CO2, or were immediately cryopreserved at -80 ̊C in either human serum
albumin or foetal calf serum with 10% DMSO, then defrosted at 37 ̊C into RPMI with 10% human
serum and all cells stimulated the next day for 4 hours and stained for intracellular markers.
Results
Typical plots showing direct comparisons of staining are shown in Figure 2.13. Human Treg
frequencies (these cells were not stimulated) were well preserved despite cryopreservation, but were
superior with HSA compared with FCS, as expected (Disis, dela Rosa et al. 2006). In a large
comparison (n=23) of the effects of cryopreservation on PBMC there was no significant effect on Treg
frequencies (Figure 2.14 A, Table 2.9), but there was a 56% fall in observed median frequencies of
TH17 cells (P=0.01) and a 62% fall in TH1 cells (P=0.02). Changes in TH2 cell frequencies were
smaller and not significant (P=0.1).
Timothy SC Hinks 2. Materials and Methods
75
Figure 2.13 Intracellular cytokine staining in fresh and cryopreserved cells
Intracellular cytokine staining in PBMC compared between cells which have been rested overnight in
culture and those which have been cryopreserved in human serum albumin or foetal calf serum, then
defrosted and stimulated. Plots show frequencies of TH17 and TH1 cells (A), TH2 cells (B) and Treg
(C) as a percentage of live CD4+ T cells.
Timothy SC Hinks 2. Materials and Methods
76
Table 2.9 Median frequencies of T cell subsets assessed by intracellular staining with or
without prior cryopreservation, in sputum and blood.
PBMC, n=23 Sputum, n=3
Fresh Cryopreserved P Fresh Cryopreserved P
Th17 0.55 0.24 0.01 6.2 1.4 0.1
Th1 6.3 2.4 0.02 4.8 1.5 0.1
Th2 0.15 0.10 0.8 0.50 0.20 0.6
Treg 4.4 4.1 0.3 7.1 8.5 0.8
Table shows median T cell frequencies as percentage of CD4+ T cells. P values are for paired T tests
I performed the same comparison with a much smaller (n=3) set of sputum samples (Figure 2.15 B,
Table 2.9). Due to the small sample size differences were not statistically significant, but the trends
were similar to those in blood, with no loss of Treg, but a 60-75% fall in frequencies of stimulated
TH17, TH1 and TH2 cells.
Timothy SC Hinks 2. Materials and Methods
77
Figure 2.14 The effect of cryopreservation on measurement of specific T cell subsets
Median frequencies of T cells in major T cell subsets as measured by intracellular staining in samples
of PBMC (A) and sputum (B) which have either been rested overnight in cell culture or have been first
cryopreserved at -80 ̊C or -196 ̊C in human serum albumin with 10% DMSO then defrosted at 37 ̊C
and rested overnight in tissue culture before stimulation with PMA and ionomycin for 4-5 hours in the
presence of monensin. Treg samples were not stimulated.
Conclusion
Cryopreservation has a negligible effect on Treg frequencies, but causes a 33-62% reduction in
frequencies of TH17, TH1 and TH2cells. For this reason cryopreservation was only used for PBMC
Timothy SC Hinks 2. Materials and Methods
78
from the longitudinal study where large numbers samples were expected from large numbers of
subjects at eight different time points and where immediate analysis of fresh samples daily, over
many months would not be feasible. It was anticipated that any loss of ‘signal’ from these immune
assays would be compensated for by large sample numbers.
In the case of sputum samples these data suggest that the effect may be even more dramatic and
furthermore many fewer samples were expected. Therefore every effort was made to process every
sputum sample immediately, 7 days a week, for the duration of the trial. Results from fresh and
cryopreserved samples were never included in the same analysis.
Choice of Golgi blocking agent for cryopreserved samples
I have observed lower frequencies of TH1 and TH17 cells after cryopreservation and thawing of PBMC,
than in fresh PBMC. One possible explanation for this, suggested by comparison with other workers,
was the choice of Golgi blocking agent. Monensin and brefeldin A are inhibitors of intracellular protein
transport. Their addition to cell cultures during the last hours of in vitro activation of cells results in
enhanced detection of intracellular cytokines. I therefore conducted some experiments to determine
which agent might provide the optimal performance in previously cryopreserved cells.
Monensin is an antiprotozoal agent produced by Streptomyces cinnamonensis(O'Neil-Andersen and
Lawrence 2002). It is a carboxylic sodium ionophore, which inhibits trans-Golgi function (Nylander and
Kalies 1999; O'Neil-Andersen and Lawrence 2002). by disrupting intracellular Na+ and H+ gradients it
exerts its greatest effects on the regions of the Golgi apparatus that are associated with the final
stages of secretory vesicle maturation(Mollenhauer, Morre et al. 1990; O'Neil-Andersen and
Lawrence 2002), inducing radical slowing of newly synthesised proteins, proteoglycans and plasma-
membrane glycoproteins, inhibiting endocytosis and thereby stopping protein recycling(Karlsson and
Nassberger 1995).
Brefeldin A is a naturally occurring macrocyclic lactone antibiotic, produced by a variety of fungi,
including Penicillium brefeldianum and is synthesized from palmitate (O'Neil-Andersen and Lawrence
2002). It has a number of cellular effects, including inhibition of protein transport between the
endoplasmic reticulum and the Golgi (Nylander and Kalies 1999) and transport from the trans-Golgi
compartment to the cell surface (Karlsson and Nassberger 1995).
Several publications compare brefeldin and monensin directly in fresh cells and in general suggest
that brefeldin A is a more potent, effective and less toxic inhibitor of cytokine secretion than
monensin(Nylander and Kalies 1999; Schuerwegh, Stevens et al. 2001; O'Neil-Andersen and
Lawrence 2002). There is higher spontaneous intracellular production of IL1β, IL6 and TNFα
(Schuerwegh, Stevens et al. 2001), and IFN-y (Caraher, Parenteau et al. 2000) with brefeldin than
monensin, but no difference in stimulated cells.
Timothy SC Hinks 2. Materials and Methods
79
Moreover three separate reports all noted slightly lower viability with monensin(Nylander and Kalies
1999; Schuerwegh, Stevens et al. 2001), and there has been a suggestion that monensin may
differentially kill CD4- cells, increasing the relative frequency of CD4+ cells(Nylander and Kalies
1999). Also of relevance to this work, CD4 down-regulation may be pronounced with brefeldin A
(O'Neil-Andersen and Lawrence 2002). Interestingly whilst one might consider adding both monensin
and brefeldin in combination, Bueno found Brefeldin A alone (10ug/ml) was superior to the
combination of brefeldin A and monensin, as it was frequently associated with both a higher
percentage of cytokine-positive cells and greater amounts of detectable cytokines per cell(Bueno,
Almeida et al. 2001). There is little published literature on the choice of agent in the context of
cryopreserved cells.
I therefore conducted several paired comparisons of brefeldin A (at a final concentration of 3.0 μg/ml)
and monensin (2.0 μM) on defrosted cryopreserved cells, using a total of 20 different samples of
cryopreserved PBMC stimulated for 5 hours in the presence of PMA and ionomycin.
Results
Results are presented in Figure 2.15. Mean frequencies of IFN-γ-producing (TH1), IL-17-producing
(TH17) and IL13-producing (TH2) cells tended to be higher with monensin than with brefeldin A and
this difference was significant for TH1 cells (P=0.035).
Timothy SC Hinks 2. Materials and Methods
80
Figure 2.15 Comparison of two inhibitors of Golgi function
Paired comparisons of frequencies of TH17, TH1 and TH2 cells in previously cryopreserved PBMC
stimulated with PMA and ionomycin for 5 hours in the presence of either monensin (2.0 μM) or
brefeldin A (3.0 μg/ml). p values represent paired t tests, n=23.
Conclusion
These experiments suggested that with my protocol, monensin gave better results than brefeldin A in
previously cryopreserved cells. Therefore and for consistency with my colleagues’ prior work and my
other work in fresh tissue I chose to continue using monensin rather than brefeldin.
Timothy SC Hinks 2. Materials and Methods
81
Selection and titration of antibodies
To enumerate the particular T cell subsets of interest to me it was necessary to construct my own
panels of fluorescent conjugated antibodies for flow cytometry.
Selection of antibodies was influenced partly by existing protocols within the group, (Ganesan 2010)
but I developed these into my own antibody panels according to specific needs, with the aim of
optimising performance on small tissue samples by identifying the maximum number of T cell subsets
within the minimum number of tubes. CD3-PE-Alexa 610 (MHCD0322, Caltag) performed poorly in
tissue compared with acceptable staining in PBMC, so was replaced by CD3-PE-Cy7 (SK7, BD). Treg
were defined by FOXP3 alone as a single marker required less sample and I did not attempt
functional experiments. To maximise sensitivity for intracellular stains IFN-γ-PE-Cy7 was replaced
with IFN-γ-APC as this a brighter fluorochrome.
Antibodies were titrated in blood to determine their optimal staining concentrations (Figure 2.16).
Surface stains could generally be used at high dilutions in blood, but at least double the concentration
was used for tissue samples due to the higher backgrounds observed. Intracellular stains were used
at the same concentrations in blood and tissue.
Timothy SC Hinks 2. Materials and Methods
82
Timothy SC Hinks 2. Materials and Methods
83
Figure 2.16 Titration of antibodies
Titration of antibodies to determine optimal concentration for use in the project. Titrations were
performed in PBMC. Final concentrations chosen are shown in Table 2.4 and 2.5 A. γδ TCR-FITC
(555748, clone MOPC-21, BD) B. CD3-PE-Cy7 (557851, clone SK7, BD) C. IFN-γ-APC (17-7319-82,
clone 4S.B3, eBioscience) D. CD8-APC-Cy7 (348813, clone SK7, BD) E. CD8-APC-eFlour®780 (47-
0087-41, clone SK7, eBioscience) showing very poor discrimination between positive and negative
CD8 cells. F. CD4-PerCP-Cy5.5 (552838, clone L200, eBioscience).
Timothy SC Hinks 2. Materials and Methods
84
Determination of optimum period of stimulation for MAIT cell
intracellular cytokine secretion
To enable me to analyse the phenotype of MAIT cells it was necessary to determine the optimum
period of cytokine stimulation, in particular to ensure that if a cytokine was not detected this was truly
due to the absence of the cytokine, rather than poor performance of the assay. It was conceivable
that IL-17 secretion in MAIT cells might not follow the same time course as that seen with CD4+ TH17
cells. It was also noted that cell surface marker expression is significantly affected by stimulation,
which might prove particularly problematic for cells selected on their specific surface marker
phenotype. I therefore performed two time-course experiments to determine the period for optimal IL-
17 expression and preservation of surface phenotype.
Methods
PBMC were obtained from a healthy control then cultured for 48 hours in AIMV media at 37 ̊C, 5%
CO2. 25 ng/ml PMA and ionomycin 500 ng/ml were added for 0, 4, 6, 12, 24 or 48 hours. Monensin (2
μM) were added only for last 2 hours of all experiments to ensure the effect of its toxicity was
equivalent in all samples. In a second experiment I added monensin for 5 hours, for consistency with
my standard protocol and also performed a timecourse to characterise the toxicity of monensin alone
over time.
Results
Results are shown in Figure 2.17. The following observations can be made. Firstly cell viability drops
with stimulation from nearly 100% initially to 71% at 4 hours and 48% at 6 hours (A). It then increases,
presumably due to cell proliferation. This favours the use of shorter stimulation times.
Timothy SC Hinks 2. Materials and Methods
85
Timothy SC Hinks 2. Materials and Methods
86
Figure 2.17 Determination of optimum period of stimulation for MAIT cell intracellular cytokine
secretion
A. Cell viability – measured by exclusion of Violet LIVE/DEAD Fixable – and frequencies of viable
TCR Vα7.2 cells, CD161+ T cells, and MAIT (Vα7.2+CD161+) cells in samples which have been
stimulated with PMA and ionomycin for 0 to 48 hours, showing the effects of cell toxicity, receptor
down-regulation and also new cell proliferation.
B. Changes over time in measured frequencies of T cells expressing IL-17, IFNγ, and TNFα
expressed as a percentage of live CD8- or CD4+ or MAIT cells depending on duration of stimulation.
Frequencies of TH17 cells are shown again in greater detail at bottom right.
C. The same time-course pattern is observed for secretion of IL-17 in TH17 cells (left) as with MAIT
cells (right).
Timothy SC Hinks 2. Materials and Methods
87
D. In the same experiment some cells were maintained in culture for 48 hours in the absence of
stimulation, but with the addition of monensin for 0 to 48 hours. After 48 hours monensin causes
significant cell death. (Note y axis does not start at 0).
Secondly TCR Vα7.2 and CD161 both down regulate rapidly over first 6 hours (A). Again this favours
shorter experiments, but it should be noted that even by 4 hours apparent MAIT cell frequencies have
fallen by 2/3.
Thirdly (B) TNFα expression is rapid and robust, peaking at 12 hours, but reaching 90% of this peak
value by 4 hours. IFN-γ secretion is also rapid and robust, though somewhat slower peaking at 12
hours but reaching only 42% of its peak value by 4 hours. Likewise IL-17 secretion is also rapid and
robust, peaking at 12 hours, but reaching only 31% of its peak value by 4 hours. IL-13 production is
minimal and bimodally distributed with peaks at 12 hours and then a larger peak at 48 hours, but near
the limit of detection.
Fourthly a key observation is that the time-course for IL-17 secretion by MAIT cells specifically was
identical to that by CD3+8- cells generally (C). This would support the use of standard published
protocols widely used in T cell research – 4 to 5hrs stimulation with PMA/ionomycin, in the presence
of monensin - in the investigation of IL-17 secretion by MAIT cells.
When the experiment was repeated, longer use of monensin (5 hours) was associated with greater
cell death (not shown). Used alone monensin does not cause significant toxicity until over 24 hours,
(D) but in combination with PMA and ionomycin cell viability was very low in this experiment, possibly
due to synergistic toxicity. Otherwise the cytokine secretion time-courses were similar.
Conclusions
Cell viability falls progressively with time during in the presence of PMA, ionomycin and monensin. My
data support a standard 4-5 hours stimulation with PMA, ionomycin and monensin for my work. Due
to the practicalities of differences in tissue handling slightly different durations were used for different
tissues, as outlined in chapter 2. Surface CD161 and Vα7.2 are rapidly downregulated with by this
stimulation and combined with the effects on cell viability I have shown that it will not be possible to
first stimulate cells then sort MAIT cells by surface phenotype. Instead characterisation of MAIT cell
function requires sorting of unstimulated cells in the first instance, or the establishment of ex-vivo cell
lines (clones), which is the avenue I chose to pursue.
Timothy SC Hinks 2. Materials and Methods
88
Optimisation and validation of RNA extraction method
To obtain further detailed phenotypic characterisation of T cell subsets I aimed to supplement flow
cytometry data with measurement of cytokine production and T cell transcription factors at the mRNA
level by RT-qPCR on sorted T cells and epithelial cells. An obstacle to this is that to perform
intracellular cytokine staining the samples undergo fixation-permeabilisation (30 mins in eBioscience
intracellular permeabilisation buffer containing formaldehyde). This cross-links nucleic acids to
proteins, inhibiting RNA extraction(Masuda, Ohnishi et al. 1999). Theoretically this can be overcome
by using a proteinase-K digestion step to reverse the cross linking (Masuda, Ohnishi et al. 1999).
Such methods have been used for successful RNA extraction from archived formalin-fixed paraffin
embedded (FFPE) histological tissue samples in combination with laser capture micro-
dissection(Godfrey, Kim et al. 2000; Lehmann and Kreipe 2001).
To determine whether this technique could be applied to fix-permeablised samples I compared RNA
extraction and RT-qPCR on two housekeeping genes - YWHAZ and β2microglobulin (β2M) - from
unfixed PBMC using the Stratagene Absolutely RNA Microprep RNA extraction kit (400805, Agilent),
with fix-permeabilised PBMC extracted using the Stratagene Absolutely RNA FFPE kit (400811,
Agilent) involving a proteinase-K digestion step.
Method
In two separate experiments fresh PBMC were obtained, resuspended in PBS or fixation-
permeabilisation buffer for 30 minutes, then diluted in 10-fold steps across the range 10 to 600,000
cells in PBS or perm-wash buffer respectively. Cells were then centrifuged at 400g for 5 mins,
resuspended in 100 μl lysis buffer (Agilent) with 0.7 μl β-ME or proteinase K digestion buffer (Agilent)
respectively, frozen to -80 ̊C, then defrosted and RNA extracted according to the manufacturer’s
protocols. RNA extraction form unfixed samples is described in the methods chapter. Fixed samples
were incubated with 10 μl proteinase K (final concentration 1.8 mg/ml) for 3.5 hours at 55 ̊C. Next
0.875 μl β-ME and 125 μl RNA binding buffer were added to each sample on ice, vortex homogenised
and diluted with 235 μl of sulfolane 90% (v/v) in RNase-free water, vortexed and RNA extracted using
the RNA binding spin-caps according to the manufacturer’s protocol.
RNA was quantified by spectrophotometer and reverse transcribed the same day using the Precision
nanoScriptTM kit (Primer Design). Housekeeping genes and a potential gene of interest – FOXP3 -
were quantified in duplicate by RT-qPCR using PerfectProbe primers (Primer Design).
Results
Optimal PCR signal was obtained for the housekeeping genes from 60,000 to 100,000 unfixed cells
(Figure 2.18 A and B). Below, or above this range sensitivity fell, with higher CT values observed at
600,000 cells, which may be due to saturation of the spin cup matrix, as the manufacturer’s report it is
optimised for 1 cell to 5x105 cells. (AgilentTechnologies 2008) There was virtually no detectable signal
from fix-permeabilised cells at any concentration, with CT values near 40 cycles (limit of detection) for
Timothy SC Hinks 2. Materials and Methods
89
the housekeeping gene and no detection of a gene of interest (FOXP3). The threshold for FOXP3
detection is expected to be higher as the gene is expressed predominantly in T cells (a small
proportion of PBMC) and at lower copy number than these abundant housekeeping genes and the
primer pair spans a longer fragment (154bp) than β2M (114bp) or YWHAZ (120bp). It is
recommended that FFPE is used with fragments less than 100bp (AgilentTechnologies 2008).
Figure 2.18 Comparison of RT-qPCR on fixed and unfixed T cells
Comparison of cycle threshold (CT) in 2 experiments in which RT-qPCR was used on either fresh
PBMC or PBMC which had first been formalin fixed and permeabilised. A range of starting cell
numbers were used. Unfixed cells were processed using the Agilent Microprep kit, whilst fixed cells
were processed with the Agilent FFPE kit which included a 3 hour incubation step in proteinase K.
Cells were reverse transcribed and quantified by RT-qPCR using PerfectProbe primers.
In a third experiment I tried to increase the RNA signal by using a variety of experimental conditions:
use of the Agilent Nanoprep kit which uses a smaller spin-cup membrane, optimised for 1 to 1x104
cells) compared with the Agilent Microprep kit (optimised for 1 to 5x105 cells); comparison of
proteinase K digestion for a standard 3 hours with shorter digestion for only 1 hour to minimise RNA
Timothy SC Hinks 2. Materials and Methods
90
degradation; and comparison of 1:5 and 1:10 dilutions of cDNA prior to the PCR step, as there is a
trade-off between amount of template and amount of reaction inhibitors. Again I obtained good RNA
signal from unfixed cells using the Nanoprep kit (CT 27.65 at 1:5 dilution, CT 28.15 at 1:10 dilution of
cDNA) in duplicate, whilst I obtained virtually no signal (CT ≥38.45) under any condition using fixed
cells (data not shown).
Conclusion
These data show that it is not practicable to use low numbers of sorted fix-premeabilised cells for
PCR, but that using unfixed cells, useable data can be obtained from 10,000 cells using the Nanoprep
kit at dilutions of 1:5 or 1:10 of cDNA.
As a result of this work I elected instead to divide each sample into 1/3 for immediate surface staining
and cell sorting and 2/3 for overnight culture and intracellular staining for data acquisition only.
Moreover, rather than sort epithelial cells cytometrically using antibody to epithelial cell adhesion
molecule (EpCAM, CD326), I have taken fresh epithelial cells obtained from bronchial brushings and
frozen directly in RNA lysis buffer.
Deep sequencing of the metagenome
Unprocessed samples of sputum and protected BAL were frozen and shipped to the Virgin Laboratory
(Department of Pathology and Immunology, Washington School of Medicine St Louis, MO, USA)
where they were passed through a 24 μm filter and the metagenome sequenced by pyrosequencing
using the Roche/454 next-generation sequencing platform (454 Life Sciences, Branford, CT, USA).
Data were analysed using the VirusHunter analysis pipeline (Zhao) in which microbial sequences
were identified on the basis of BLAST alignments and the taxonomic classification of the reference
sequences to which a read is aligned.
Microarray
To measure whole transcriptome gene expression, pure populations of T cells were sorted directly
into 100 μl Agilent lysis buffer with 0.7 μl of β-ME and homogenised, frozen and shipped at -80 ̊C to
Janssen Research & Development (Springhouse, PA, USA) for further analysis as part of a
collaboration between the company and my supervisor. RNA was extracted using the Absolutely RNA
Nanoprep Kit, reverse transcribed and amplified by in vitro transcription with the Ovation Pico WTA
System V2 (NuGEN Technologies, San Carlos, USA) and gene expression measured with the
GeneChip® HT HG-U133+ PM Array Plate (Affymetrix, Santa Clara, USA) on an Agilent GeneArray
Scanner. Samples which passed hybridization signal intensity threshold were robust multi-array
average (RMA)/log2 transformed. Additional quality control (principal component analysis, correlation
and median absolute deviation score (MADscore)) was performed in ArrayStudio v6.1 software
(Omicsoft Corporation, Cary, NC).
Timothy SC Hinks 2. Materials and Methods
91
Statistical Analysis
Data elaboration and preparation for analysis
Data in the cross sectional study were presented as median (interquartile range (IQR)) and frequency
(percentage), unless stated otherwise. Normality of the quantitative (numeric) data distributions was
tested by the Shapiro-Wilk test. If the data were not normally distributed they were logarithmically
transformed.
Cross sectional study
Data were analysed by descriptive and exploratory statistical methods to compare relationships
between variables. The data were analysed using Student’s t tests (two groups’ comparison) and
where multiple groups were compared by one-way analysis of variance (ANOVA), the post hoc
Dunnett’s test was also applied (healthy control as reference category). If the groups were ranked
according to disease severity, data were tested for linear trend using polynomial contrasts method.
If data were not normally distributed, or the logarithmic transformation generated a loss of many zero
values, non-parametric tests were used. Alternatively, a minimal constant of 0.01 or 0.001 was added
to each of the values to allow for a logarithmic transformation. The Mann-Whitney-U test was used to
compare two groups, while multiple groups were compared by one-way ANOVA using the Kruskal-
Wallis method (with post hoc Dunn-bonferoni’s test applied at the p-values of the significant
differneces) (Dunn 1964) Linear trend across ranked groups was tested using the pairwise
Jonckheere-Terpstra test.
If cyclical patterns were suspected in the variations over time (e.g., seasonality in MAIT cell
frequencies), non-linear regression was performed on the log transformed data using a standard sine
function, as follows:
Ln (MAIT frequency)=0.0795+0.6024*sin((2π*[Seasonal quarter]/4)+78.27)
The modeling of ELISA and MSD data was performed by standard regression equation with a 5-
parameter curve fit.
Exploratory Analyses of Relationships between variables
Additional explorative (hypothesis-generating) investigation was performed to identify possible
associations among the demographic, behavioural and clinical parameters in the asthma patients,
including the type of asthma (mild, moderate and severe). We used factor analysis (principal
component method, or analysis, PCA) by varimax rotation with Kaiser normalisation and generation of
scree plots (minimum 80% explained variance by the identified components). Only the variables with
a rotated component matrix score >0.50 were considered due to limited sample size (Field 2000). To
further check the robustness of identified principal components (clusters), a reliability analysis with
computation of Cronbach’s alpha coefficient was applied (>0.70 was assumed as acceptable) (Kline
Timothy SC Hinks 2. Materials and Methods
92
1999). Relationships between pairs of continuous variables were also explored using Spearman’s
rank correlations.
The data for the longitudinal study were collected and analysed as paired observations per patient
over time, for each individual variable, or as averaged, time-series data. Parametric and non-
parametric paired tests were used, except for the data on fresh sputum and PBMC as very few
samples were present. Analyses were essentially exploratory and descriptive. Analyses were planned
a priori including use of ANOVA and t tests to compare baseline (T0) data with data from the time-
point of peak difference.
In particular, a number of characteristics (e.g., TH17 frequency) were measured over time in the
groups of patients with rhIFN-β1α and placebo. Repeated measures ANOVA indicated a marginally
significant univariate difference between the groups, but due to the lack of normal distribution, a pre-
planned analysis further computed the individual areas under the curve (AUC) as summary measures
of the overall dynamics for each variable. The AUCs were compared between the two groups by t test
according to the method of Matthews (Matthews, Altman et al. 1990).
Timothy SC Hinks 2. Materials and Methods
93
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
95
CHAPTER 3
CD4+ T cell phenotypes in asthma Let no man think or maintain that anyone can search too far or be too well studied in the book of
God’s words or in the book of God’s works; rather let all endeavour an endless progress or
proficience in both.3
3 On page ii of The Origin of Species, Charles Darwin FRS (1809-1882) quoted these words
from Francis Bacon’s Advancement of Learning (1605). Darwin, C. (1859). On The Origin Of
Species By Means Of Natural Selection. London, John Murray.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
96
Introduction
For over a decade many workers have hypothesised a significant role for IL-17 and the TH17 cell
subset in the pathogenesis of human asthma(Molet, Hamid et al. 2001) on the basis of human genetic
associations(Hizawa, Kawaguchi et al. 2006; Kawaguchi, Takahashi et al. 2006; Chen, Deng et al.
2010; Lluis, Schedel et al. 2011), murine data (Park, Li et al. 2005; Schnyder-Candrian, Togbe et al.
2006; Fujiwara, Hirose et al. 2007; McKinley, Alcorn et al. 2008) and reports of IL-17 protein and
mRNA in airway samples(Molet, Hamid et al. 2001; Barczyk, Pierzchala et al. 2003; Chakir, Shannon
et al. 2003; Bullens, Truyen et al. 2006). Despite this there remains a pauctiy of robust human data
and there are no studies which have investigated airway TH17 cells in human asthma. I therefore
sought to scrutinise these hypotheses by conducting a comprehensive review of TH17 cells in the
context of other major CD4+ T cell subsets in peripheral blood, sputum, BAL and bronchial biopsies
from subjects with a range of asthma phenotypes and healthy controls during periods of clinical
stability. This chapter describes my findings of this crosss-sectional study, including an analysis of
serum and airway IL-17 protein in the context of other major T cell cytokines complemented by flow
cytometry data on different T-helper cell subsets in blood, sputum, BAL and tissues.
Results and comments
Study population
23 healthy subjects and 53 asthmatics (14 mild, steroid-naïve, 17 moderate, treated with low dose
inhaled corticosteroids and 22 severe, treated with oral or high dose inhaled corticosteroids) were
studied. All had stable symptoms for at least 6 weeks prior to clinical sampling. The study design is
shown in Figure 2.1 and clinical characteristics of the study participants are shown in table 3.1.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
97
Table 3.1 Demographic and clinical characteristics of cross sectional cohort for CD4+ and
CD8+ T cell analysis
n 23 14 17 22DemographicsSex (M/F) 14 / 9 8 / 6 8 / 9 8 / 14Age (median [range], years) 28 (20-65) 26 (21-64) 35 (21-56) 53 (23-67)Pulmonary function
FEV1 (% predicted) 108 (104-113) 88 (85-101) 99 (86-109) 65 (49-82)FEV1 reversibility (%) 3.6 (1.8-7.9) 14 (9.9-19) 12 (6.7-19) 13 (2.6-25)PEFR (% predicted) 108 (97-116) 97 (89-108) 95 (88-99) 70 (53-82)PEFR variability (%) 15 (N/A) 17 (10-27) 22 (16-34) 17 (12-24)PD20 (mg methacholine) 0.18 (0.044-0.48) 0.25 (0.057-0.58)
Exhaled nitric oxide (ppb, at 50 L/s) 16 (11-21) 56 (30-110) 27 (14-49) 20 (13-38)ClinicalAtopy (Skin prick positive, Y/N) 0 / 23 14 / 0 15 / 2 15 / 7
No. of skin prick allergens positive 0 (N/A) 6 (4-7) 3 (2.5-5) 3.5 (0-5.3)
Peripheral eosinophil count (109/L) 0.1 (0.1-0.2) 0.1 (0.1-0.6) 0.2 (0.2-0.3) 0.2 (0.1-0.3)Total IgE (iu/ml) 32 (9.4-62) 173 (62-457) 119 (25-188) 84 (31-669)
Body mass index (kg/m2) 24.5 (22.3-28.2) 23.6 (22.5-26.7) 25 (22.7-31.5) 31 (27.1-40.9)Smoking status
Never 21 13 15 17Former (Mean pack years) 2 (4) 1 (5) 2 (1.8) 4 (26)Current (Mean pack years) 0 0 0 1 (49)
Duration of asthma (years) N/A 18 (15-25) 22 (10-25) 36 (21-49)ACQ score N/A 0.65 (0.43-1.3) 1.3 (0.75-1.8) 2.8 (2.2-3.5)GINA level of control
Controlled N/A 7 (50) 3 (18) 0 (0)Partly controlled N/A 6 (43) 11 (65) 2 (9.5)Uncontrolled N/A 1 (7.1) 3 (18) 19 (90)
TreatmentInhaled steroids No No Yes Yes
Dose (equivalent mcg BDP) N/A N/A 400 (400-900) 1600 (1280-2000)Maintenance oral steroids (Y,N) No No No 6 / 16
Mean dose if taken (mg prednisolone/day) 11Long acting β agonist (Y/N) No No 8 / 9 22 / 0Leukotriene receptor antagonist (Y/N) No No No 15 / 7Step on BTS treatment algorithm N/A 1 2 - 3 4 - 5
Inflammatory subtype (n, %)Neutrophilic 4 (25) 2 (14) 2 (14) 10 (48)Eosinophilic 1 (6.3) 3 (21) 3 (21) 6 (29)Mixed granulocytic 0 (0) 0 (0) 0 (0) 1 (4.8)Paucigranulocytic 11 (69) 7 (50) 9 (64) 4 (19)
Sputum cell differential (%)Macrophages 52 (31-66) 45 (34-62) 53 (31-65) 30 (19-43)Neutrophils 31 (11-65) 35 (22-58) 33 (16-56) 61 (32-76)Epithelial 3.6 (2.0-24) 4.1 (0.83-11) 3.8 (1.1-16) 2.9 (0-7.8)Eosinophils 0.38 (0-0.94) 1.5 (0.75-1.8) 1 (0.38-1.8) 0.69 (0-6.1)Lymphocytes 0.1 (0-0.75) 0.3 (0-0.75) 0 (0-0.63) 0.0 (0-0.25)
BAL cell differential (%)Macrophages 84 (74-89) 70 (60-80) 81 (73-89) 72 (46-94)Neutrophils 2.5 (1.0-5.9) 2.5 (1.6-4.8) 3.5 (1.8-6.4) 6.5 (1.4-29)Epithelial 9.9 (3.9-18) 21 (13-35) 11 (5.6-19) 8.7 (3.3-11)Eosinophils 0.25 (0.0-0.56) 2.0 (0.75-3.6) 1.0 (0-3.0) 0.1 (0-1.6)Lymphocytes 1.4 (0.94-2.4) 1.5 (0.38-3.0) 1.3 (0.5-2.3) 1 (0-1.6)
Relevant comorbidities (n, %)Allergic rhinitis 0 (0) 11 (79) 8 (47) 10 (46)Nasal Polyps 0 (0) 0 (0) 1 (5.9) 5 (23)Eczema 1 (13) 7 (50) 5 (29) 4 (19)Bronchiectasis (history or CT) 0 (0) 0 (0) 1 (5.9) 1 (4.5)
Values are medians with interquartile ranges, unless stated otherwise. N/A: not available.Inflammatory subtype is based on sputum differentials using cut-points as per Simpson, J. L., R. Scott, et al. (2006). Respirology 11(1): 54-61 (neutrophilic: >61% neutrophils, eosinophilic: >3%). Percentages are of those with valid data.ACQ, asthma control questionnaire; BDP, beclometasone dipropionate; BTS, British Thoracic Society; CT, computed tomogram; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; GINA, Global Initiative for Asthma; PEFR, peak expiratory flow rate; PD20, provocative dose 20.
Healthy controls Mild asthma Moderate asthma Severe asthma
Negative Not done
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
98
Measurement of IL-17 protein by enzyme-linked immunosorbent assay (ELISA)
A previous study has observed increased IL-17 protein levels in sputum samples of asthmatics using
ELISA(Barczyk, Pierzchala et al. 2003). Therefore, to determine whether IL-17 protein could indeed
be measured in respiratory specimens, I used an ELISA assay (88-7976, eBioscience) to detect IL-17
in a variety of different samples. Samples tested in duplicate included supernatants from sputum
(n=15) and BAL (n=15) obtained during periods of clinical stability, and also supernatants from
bronchial biopsies with and without ex vivo allergen challenge and lung parenchymal samples with or
without ex vivo challenge with live X31 influenza virus. The assay produced a good standard curve
with good replicates over the range 4-500 pg/ml (Figure 3.1 A), but IL-17 was not detected in any of
the samples tested, suggesting it is either produced at very low abundance in these samples or very
unstable, and that ELISA is not appropriate for this application.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
99
Figure 3.1 ELISA standard curves
Standard curves, using 5 parameter curve-fit, for ELISA experiments performed.
A. IL-17 showing accuracy over the range 4 to 500 pg/ml. (eBioscience 88-7976) Samples assayed
included supernatants from sputum and BAL during clinical stability, bronchial biopsy supernatants
with and without ex vivo allergen challenge and lung parenchymal samples with or without ex vivo
challenge with live influenza virus.
B. Standard curve of serum IgE measured by ELISA showing accuracy over the range 7.8 to 500
ng/ml. (eBioscience BMS2097)
C. Serum IgE levels in healthy and asthmatic subjects. Differnces are compared by Kruskal-Wallis
ANOVA with post hoc Dunn’s.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
100
Measurement of serum IgE
Serum IgE was measured by ELISA. The limit of detection (LOD) was 7.8 ng/ml and the results are
presented in Figure 3.1B and included in table 3.1.
Detection of cytokines by electrochemiluminescence (MSD)
Using standard ELISA with a limit of detection of 4 pg/ml, I was unable to detect IL-17. Of note, others
have since reported that IL-17 levels are typically <4 pg/ml in serum(Zhao, Yang et al. 2010) and
frequently <2 pg/ml in sputum(Doe, Bafadhel et al. 2010). Therefore I chose to measure serum and
airway cytokines using the more sensitive technique of multiplexed enzyme-linked
electrochemiluminescent assay using the Meso Scale Discovery (MSD) platform.
According to the MSD manufacturer’s reports 10 mM concentrations of dithiothreitol (DTT), a strong
reducing agent used for processing sputum samples, denature antibodies thereby reducing assay
sensitivity, while this effect has been found to be minimal at 1 mM by researchers (Yvonne Clements,
personal communication). To achieve a compromise between dilution effects and antibody
denaturation I decided to use dithioerythritol (DTE) at 5 mM. I directly compared assay sensitivity for a
range of T cell cytokines diluted either in proprietary diluent or in a 1:1 mix of DTE with proprietary
diluent. At this concentration I observed very close agreement for all cytokines tested across a wide
range of concentrations from 0.61 to 2500 pg/ml (Figure 3.2 A, Table 3.2).
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
101
Figure 3.2 Validation of MSD in sputum
Measurement of cytokines by Meso Scale Discovery multiplex ELISA
A. To determine the most appropriate buffer for MSD when analysing sputum samples standard
curves were prepared in proprietary diluent and also in a 1:1 mix of DTE with proprietary diluent. Very
close agreement was observed across a wide range of concentrations from 0.61 to 2500 pg/ml.
B. Spiking recovery from sputum was tested for each cytokine in duplicate wells, in samples from
three different subjects, at three different concentrations: 10, 100 and 1000 pg/ml.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
102
Table 3.2 Percentage of cytokine measured in DTE/diluent 1:1 mix compared with that in
proprietary diluent alone across the lower dynamic range (0-39 pg/ml)
Cytokine IFN- IL-10 IL-12 p70 IL-13 IL-2 IL-4 IL-5 IL-17
Detection in presence of
DTE (% of detection in
proprietary diluent alone)
92 104 67 75 91 78 87 93
Barczyk reported a three- to four-fold decrease in IL-17 measured by ELISA when using 0.05% DTT
(Barczyk, Pierzchala et al. 2003). By contrast my findings suggest that sputum processing with 5 mM
(0.1% w/v) DTE does not impair sensitivity of the MSD assay.
Next I tested spiking recovery in sputum samples. Three samples were tested in duplicate wells at 3
different concentrations (10 pg/ml, 100 pg/ml and 1000 pg/ml)(Figure 3.2 B). Average spiking
recovery at the 10 pg/ml concentration was 54% but depended on the cytokine being assayed, as
shown in table 3.3.
Table 3.3 Average spiking recovery from sputum using 10 pg/ml spikes.
Cytokine IL-17 IFN- IL-2 IL-4 IL-5 IL-10 IL-12p70 IL-13
Recovery (%) 67.0 89.3 37.7 39.5 94.3 69.4 70.9 41.2
In BAL spiking recovery was generally higher with an average recovery of 66%. As BAL
samples underwent centrifugal dialysis the spike was added prior to concentration step.
Individual BAL spike recoveries are shown in table 3.4.
Table 3.4 Average spiking recovery from BAL using 10 pg/ml spikes.
Cytokine IL-17 IFN- I-L2 IL-4 IL-5 IL-10 IL-12p70 IL-13
Recovery (%) 59.1 76.0 56.2 60.6 89.4 50.6 66.5 68.0
Limits of detection for each sample type are shown in table 3.5. The use of MSD and, in the case of
BAL samples, additional concentration by centrifugal dialysis allows me to achieve threshold
sensitivities an order of magnitude different from prior literature in the field. My assay sensitivity for IL-
17 is 70 times greater than that used by Barczyk or Doe in sputum(Barczyk, Pierzchala et al. 2003;
Doe, Bafadhel et al. 2010). My threshold sensitivity for serum IL-17 of 0.147 pg/ml is at least 20 times
greater than that achieved by others with ELISA (LOD 4-15 pg/ml (Molet, Hamid et al. 2001; Zhao,
Yang et al. 2010; Bazzi, Sultan et al. 2011)) or with Luminex® (LOD 3.2 pg/ml(Zhao, Yang et al.
2010)). My combination of MSD and sample concentration makes my measurement of BAL IL-17
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
103
1000 fold more sensitive than that achieved by standard ELISA without concentration(Song, Luo et al.
2008).
Table 3.5 Effective limits of detection for cytokines measurement by MSD for each tissue
(pg/ml).
Sample Cytokine
type IL-17 IFN- IL-2 IL-4 IL-5 IL-10 IL-12 p70 IL-13
Serum 0.147 1.05 0.289 1.20 0.520 0.887 0.748 1.27
BAL* 0.00395 0.0317 0.0284 0.0730 0.0113 0.0272 0.0223 0.00197
Sputum† 0.216 0.695 0.122 0.615 0.272 0.345 0.510 0.885
*BAL samples were concentrated 50 fold. †Sputum was diluted 2 fold.
Cytokines measured by MSD in serum
Serum concentrations of the following eight cardinal T-cell cytokines were similar between asthma
and health: IL-17, IL-2, IL-10, the TH1 cytokines IFN-, IL-12p70 and the TH2 cytokines IL-4, IL-5 and
IL-13(Figure 3.3). Groups were compared by ANOVA on Ln transformed data, and by test for linear
trend across groups. Furthermore no significant differences were observed even when all asthmatic
subjects were combined (Figure 3_6 A). This finding is at odds with an observation by another group
which found plasma IL-17 levels measured by Luminex assay in 12 healthy controls to be uniformly
below the limit of detection (3.2 pg/ml), whilst in 29 subjects with allergic asthma they reported a
mean level of 12.5 pg/ml (Zhao, Yang et al. 2010). Perhaps differences are due to the differences in
method, antibodies or choice of plasma rather than serum. However, MSD is 20 times more sensitive
than Luminex, and the data-set analysed in my study is considerably larger and therefore less
susceptible to distortion of the mean value by a outliers. Furthermore, in the largest published
comparison of serum IL-17 levels to date (Bazzi, Sultan et al. 2011), no significant differences were
observed in serum IL-17 levels measured by ELISA between 100 asthmatics and 102 healthy
controls, which is fully consistent with my findings.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
104
Figure 3.3 Cytokines measured by multiplex ELISA in serum
T cell cytokines measured by multiplex ELISA using the Meso Scale Discovery (MSD) platform from
serum samples. Samples were measured in duplicate wells from 64 individuals (18 healthy controls,
12 mild, 16 moderate and 18 severe asthmatics) and are expressed in pg/ml. (A) IL-17 and IL-10. (B)
TH1 cytokines. (C) TH2 cytokines. No significant differences were observed between groups. LOD,
limit of detection (pg/ml); HC, healthy control.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
105
Cytokines measured by MSD in BAL
Interesting data were obtained on cytokine levels in BAL. As expected levels of the TH2 cytokines IL-5
and IL-13 were markedly elevated in asthma with ANOVA (P<0.0001 and P=0.02 respectively)(Figure
3.4, 3.6 C). This is consistent with the current dogma that allergic asthma is characterised by TH2
inflammation (Robinson, Hamid et al. 1992; Till, Durham et al. 1998; Larche, Robinson et al. 2003).
IL-4, a third TH2 cytokine, was not detected in BAL, consistent with our own group’s previous
experience. Furthermore, the elevation of IL-5 and IL-13 was most consistently observed in the
subgroup with mild, steroid naïve atopic asthma. Amongst the moderate and severe asthmatic
subgroups, BAL TH2 levels may have been lower in some individuals as a result of treatment (Naseer,
Minshall et al. 1997; Richards, Fernandez et al. 2000; Di Lorenzo, Pacor et al. 2002) or due to
intrinsic differences in the underlying asthma phenotype (Woodruff, Modrek et al. 2009).
Figure 3.4 Cytokines measured by multiplex ELISA in bronchoalveolar lavage
T cell cytokines measured by multiplex ELISA using the MSD platform from serum samples. Samples
were measured in duplicate wells from 59 individuals (18 healthy controls, 12 mild, 16 moderate and
13 severe asthmatics) and are expressed in pg/ml. Samples were first concentrated by centrifugal
dialysis. (A) IL-17 and IL-10. (B) TH1 cytokines. (C) TH2 cytokines. Distributions were compared by
ANOVA on Ln transformed data P values are given where P<0.05. Ln transformed data were also
tested for linear trend across groups and significant results are presented. LOD, limit of detection
(pg/ml); HC, healthy control.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
106
Figure 3.5 Cytokines measured by multiplex ELISA in sputum
T cell cytokines measured by multiplex ELISA using the MSD platform from serum samples. Samples
were measured in duplicate wells from 48 individuals (14 healthy controls, 8 mild, 12 moderate and 14
severe asthmatics) and are expressed in pg/ml. (A) IL-17 and IL-10. (B) TH1 cytokines. (C) TH2
cytokines. Distributions were compared by ANOVA on Ln transformed data P values are given where
P<0.05. Ln transformed data were also tested for linear trend across groups and significant results
are presented. LOD, limit of detection (pg/ml); HC, healthy control.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
107
Figure 3.6 Cytokines measured by multiplex ELISA compared between asthma and health
Results from Figures 3.3 to 3.5 stratified to compare health with all asthma combined. P values
represent statistically significant differences between asthma and health using unpaired t tests on ln
transformed data. After correction for multiple comparisons using P’=P*√n differences remain
significant at P<0.05 except for sputum IL-12p70.
Importantly, BAL IL-17 levels did differ between asthma subgroups (ANOVA P=0.04, Figure 3.4),
being higher in a subset of mild, steroid naïve asthmatics, although there was no overall difference
between health and asthma when all asthmatic subjects were combined (P=0.3, Figure 3.6).
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
108
Figure 3.7 Correlates of BAL IL-17 levels
The presence of allergic rhinitis is associated with elevated concentrations of IL-17 in bronchoalveolar
lavage in (A) asthmatic subjects and (B) in all subjects combined. P values are for Mann Whitney U
test. BAL IL-17 concentrations also correlate with (C) BAL eosinophil counts, with (D) serum IgE
levels nad with (E) exhaled nitric oxide. Statistics are for Spearman’s correlation.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
109
Figure 3.8 Relationship between BAL IL-17 levels and BAL epithelial cells
BAL IL-17 levels are significantly correlated with the proportion of epithelial cells present in the same
sample expressed as a percentage of total differential cell count. (A) Relationship is tested by
Spearman’s correlation (rs=0.362, P=0.007). As correlations may be misleading where there is an
outlying group the data are also presented as BAL IL-17 levels according to whether epithelial cell
counts were within the normal range, or abnormally high. (B) The upper limit of the normal range is
≤24% based on the 2.5th-97.5th percentile in my healthy controls. Mean BAL IL-17 levels are 3.6 fold
higher in subjects with abnormally high epithelial counts (Mann Whitney P=0.0008).
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
110
Figure 3.9 Correlates of airway TH2 cytokines
Associations of airway TH2 cytokine concentrations with markers of eosinophilic inflammation, tested
by Spearman’s correlations. (A) IL-13 and (B) IL-5 levels in BAL are positively correlated with exhaled
nitric oxide and eosinophils in sputum and BAL. (C) Sputum IL-5 levels correlate negatively with lung
function, and also weakly correlate positively with sputum neutrophils.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
111
Figure 3.10 Airway cytokines according to inflammatory phenotype
Airway cytokine levels stratified according to asthmatic inflammatory phenotype based on sputum cell
differentials: (A) sputum IL-17, (B) BAL IL-17, (C) BAL IL-5, (D) BAL-13. No differences are significant
(Kruskall Wallis test), but this is likely to be because of the low number of eosinophilic asthmatics
included. Note that many subjects could not expectorate and therefore were not classified according
to inflammatory subtype.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
112
Five of the six subjects with high BAL IL-17 levels had mild, steroid naïve asthma. Compared to other
mild asthmatics they were older with a mean age of 39 years versus a group mean of 25 years
(P=0.02). The one severe asthmatic individual with high BAL IL-17 was also older at 63 years of age.
High BAL IL-17 levels were also associated with the presence of allergic rhinitis (P=0.02), BAL
eosinophilia (rs=0.34, P=0.04), high serum IgE (rs=0.42, P=0.007, Figure 3.7) and a tendency towards
an eosinophilic sputum inflammatory subtype (NS) (Figure 3.10). In the light of prior literature
(Barczyk, Pierzchala et al. 2003) I found no evidence for any association between BHR and IL-17
levels in sputum (rs=-0.07, P=0.7, n=29) or BAL(rs=0.02, P=0.9, n=39).
Elevated BAL IL-17 levels may be associated with epithelial fragility
Epithelial cell derived IL-17 has recently been implicated in nasal inflammation (Semik-Orzech,
Barczyk et al. 2009; Saitoh, Kusunoki et al. 2010; Xu, Zhang et al. 2010; Jiang, Li et al. 2011; Quan,
Zhang et al. 2012). In light of the above finding of higher BAL IL-17 levels in patients with allergic
rhinitis, I wondered whether the bronchial epithelium might be an important source of BAL IL-17.
During processing of BAL samples I noted occasional individuals in whom BAL contained sheets of
bronchial epithelium. I therefore looked for an association between BAL IL-17 levels and the presence
of epithelial cells in the same BAL sample, as measured on BAL cytospins. There was indeed a
significant correlation between the number of epithelial cells present in the BAL cytospins and the
amount of BAL IL-17 (Spearman’s correlation rs=0.362, P=0.007)(Figure 3.8 A). All but one of the
subjects with high IL-17 levels clustered with very high epithelial cell counts. However, such simple
correlations may be misleading. From my data-set, I determined that the upper limit of the normal
range for BAL epithelial cell counts is ≤24% based on the 2.5th-97.5th percentile in my healthy
controls, which allowed me to dichotomise the subjects into those with normal epithelial cell
frequencies and those with abnormally high epithelial cell contamination(Figure 3.8 B). Using this
analysis it can be seen that mean BAL IL-17 levels are 3.6 fold higher in subjects with abnormally
high epithelial counts (Mann Whitney P=0.0008). Furthermore it should be noted that this cluster of
high BAL IL-17 and high epithelial cell counts comprised the same five steroid-naïve, mild asthmatics,
whilst the one high IL-17 severe asthmatic subject (413) was again the outlier with normal epithelial
cell numbers.
Several mechanism could explain these observations. First, it is conceivable that the cellular source
of the BAL IL-17 in the above subset of asthmatics is the inflamed lower airway epithelium, analogous
to the situation in the upper airway epithelium (Semik-Orzech, Barczyk et al. 2009; Saitoh, Kusunoki
et al. 2010; Xu, Zhang et al. 2010; Jiang, Li et al. 2011; Quan, Zhang et al. 2012). Indeed it has long
been recognised that inflammation of the upper airway may be intimately linked with that in the lower
airway (Mackenzie 1885) and this is suggested by the association in my data-set between allergic
rhinitis and elevated BAL IL-17. Furthermore inflammation of the lower airway epithelium is known to
cause loss of epithelial integrity with a selective loss of columnar epithelial cells and disruption to tight
junctions with loss of junctional proteins such as ZO-1 and E-cadherin (Swindle, Collins et al. 2009).
Such disruption of tight junctions can be promoted by T-cell cytokines including IL-13 and TNF-α
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
113
(Swindle, Collins et al. 2009) and IL-17 (Kebir, Kreymborg et al. 2007; Huppert, Closhen et al. 2010;
Gutowska-Owsiak, Schaupp et al. 2012; Soyka, Wawrzyniak et al. 2012) leading to epithelial fragility
which might predispose to the increased epithelial cell sloughing during bronchoalveolar lavage which
I have observed in steroid naïve asthmatics.
Cytokines measured by MSD in sputum
When measured in sputum the TH2 cytokine IL-5 was again increased in asthma (ANOVA P=0.005,
Figure 3.5), although the pattern according to disease phenotype was different from that observed in
BAL, with the highest levels correlating with the greatest disease severity (P for linear trend across
groups =0.0006). The expression pattern of other cytokines also differed between BAL and sputum,
which is typical of the experience of our group, reflecting the different cellular and protein composition
of sputum and its more proximal origin.
I observed no significant differences in sputum IL-13 orin sputum IL-17 levels (Figure 3.5) which is
contrasts with the findings of Barczyk et al. Possible explanations for these differences between
Barczyk et al. and my own findings includes study size and assay sensitivity; more than twice as
many subjects were analysed in my study (48 v 21 sputum samples), and the MSD assay used here
is 70 times more sensitive than the Luminex assay employed by Barczyk et al.
Sputum IL-2 levels differed between mild and severe asthma (ANOVA P=0.03). IL-2 is produced by
activated TH1 cells and has been found to induce bronchial hyper-reactivity in rats (Barnes,
Djukanovic et al. 2003), but the significance of my finding is not clear, particularly as there was no
overall difference between asthma and health (Figure 3.6).
Sputum IL-12p70 levels were lower in asthma than in health (P=0.02, Figure 3.5 and 3.6). IL-12 is
produced by monocytes and macrophages to promote differentiation of naïve T cells into TH1
cells(Hsieh, Macatonia et al. 1993), and can inhibit BHR and airway eosinophilia in animal
models(Barnes, Djukanovic et al. 2003). Therefore IL-12 might be expected to be deficient in subjects
with TH2 mediated allergic asthma, indeed IL-12 has been reported to be deficient in peripheral blood
in allergic asthma(Barnes, Djukanovic et al. 2003).
In summary measurement of eight cardinal T-cell cytokines in a range of tissues revealed a lack of
systemic markers of T-cell response in blood, but consistent evidence of TH2 inflammation in airway
samples. My data do not show a generalised increase of IL-17 in asthma, although they suggest that
IL-17 may be elevated in a discrete subgroup of mild, steroid naïve asthmatics, which tend to be older
with atopic, eosinophilic asthma and allergic rhinitis. These may represent a distinct endotype of
asthma. These data do not however identify a specific cellular source of IL-17 in the airways.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
114
Measurement of IL-17 in airway macrophages by RT-qPCR
Airway macrophages constitute one possible cellular source of IL-17(Song, Luo et al. 2008; Park and
Lee 2010; Reynolds, Angkasekwinai et al. 2010). I therefore measured IL-17 mRNA by RT-qPCR in
live CD45+CD3-HLADR+ sputum macrophages sorted by flow cytometry from 29 subjects comprising
10 healthy controls, 9 mild asthmatics and 10 moderate asthmatics. IL-17 mRNA was quantified in
triplicate using PerfectProbe primers for IL-17A and normalised to β2 microglobulin. IL-17 mRNA was
detected only from a single mild asthmatic subject (211) at an average cycle threshold (CT) of 37.2
compared with a CT of 14.6 for the house-keeping gene, implying very low transcript abundance (see
Figure 3.11). Although airway macrophages have been identified as a source of IL-17 in a murine
model of asthma(Song, Luo et al. 2008), my data imply that airway macrophages are not a principle
source of IL-17 in the airways in humans.
Figure 3.11 Airway macrophage expression of IL-17 mRNA
IL-17 mRNA was measured by RT-qPCR in airway macrophages. Live CD45+CD3-HLADR+ were
obtained from sputum and sorted by flow cytometry and mRNA was quantified in triplicate using
PerfectProbe primers for IL-17, and normalised to β2 microglobulin. 29 samples were tested from 10
healthy controls, 9 mild asthmatics and 10 moderate asthmatics. IL-17 mRNA was detected only from
a single subject, 211, who had mild asthma.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
115
Figure 3.12 Major CD4+ T cell subsets in asthma and health
Frequencies of T cells expressing (A) IL-17 (TH17 cells), (B) IFN-γ (TH1 cells), (C) IL-13 (TH2 cells)
and (D) FOXP3 (T reg) in PBMC, sputum, BAL and bronchial biopsies measured by intracellular
cytokine staining and flow cytometry. Results expressed as a percentage of live CD3+CD4+ T cells.
In the case of bronchial biopsies frequencies are a percentage of CD3+8- T cells. Differences are
compared by Mann-Whitney U tests and significances given where P<0.05.
healthy controls; asthmatic subjects.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
116
Cytometry of major CD4+ T cell subsets in asthma
My next objective was to enumerate key CD4+ T cell subsets in PBMC, sputum, BAL and bronchial
biopsies across a spectrum of asthma phenotypes. Flow cytometry is an ideal technique for such
work as it combines the high sensitivity and high specificity needed for the accurate detection of rare
events with the ability to determine the exact cellular origin of cytokines on a cell-by-cell
basis(Baumgarth and Roederer 2000).
Evidence of increased TH2 cell inflammation, but no differences in TH17 frequencies in asthma
Figure 3.12 shows data from 23 healthy subjects and 53 asthmatics comparing frequencies of TH17,
TH1, TH2 and Treg cells in asthma and health. Contrary to my hypotheses, I found no significant
differences in frequencies of TH17 cells between health and asthma in blood or any tissue
compartment (A). The same was true for TH1 cells (B) and these findings were in clear contrast to my
observations of an increase in TH2 cells in bronchial biopsies (C) with a median 0.36% (IQR 0.19-
1.5%) in asthma compared with 0.10% (0.025-1.3%) in health (Mann-Whitney P=0.047, n=47), with
similar trends in PBMC, sputum and BAL (NS). In addition I also observed a decrement in Treg in
BAL (D) in asthmatics at 5.3% (4.3-8.2%) compared with health (8.1% (5.6-10%) P=0.027, n=67).
These findings are analysed in greater detail in Figure 3.13 where I have stratified the asthmatic
individuals according to disease severity. Again there is clearly no evidence of differences in TH17 cell
frequencies between health or any asthma phenotype (A), whilst the differences in TH2 cell
frequencies are more apparent (C), being most strikingly elevated in mild, steroid naïve asthmatics in
PBMC (P=0.003), sputum (P=0.03) and biopsies (P=0.02) with a similar trend in BAL (NS). It can also
be seen that the deficiency in BAL Treg correlates with disease severity (P for linear trend =0.02)
being most marked in severe asthma with frequencies of 4.4% (3.1-6.1%) compared with 8.1% (5.6-
10%) in health (P=0.00).
I further analysed this evidence of a bias towards TH2 inflammation by comparing ratios of TH2 to TH1
cells in each tissue compartment (Figure 3.14). An increase in the TH2:TH1 ratio was observed in
sputum (Kruskal-Wallis P=0.01), BAL (P=0.049) and bronchial biopsies (P=0.009), with a similar
pattern in PBMC (NS). Again this TH2 bias was most marked in mild, steroid naïve asthmatics.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
117
Figure 3.13 Major CD4+ T cell subsets stratified by disease severity
Frequencies of T cells expressing (A) IL-17 (TH17 cells), (B) IFN-γ (TH1 cells), (C) IL-13 (TH2 cells)
and (D) FOXP3 (T reg) in PBMC, sputum, BAL and bronchial biopsies measured by intracellular
cytokine staining and flow cytometry. Results expressed as a percentage of live CD3+CD4+ T cells.
In the case of bronchial biopsies frequencies are a percentage of CD3+8- T cells. Differences are
compared by Kruskal-Wallis tests and significances given where P<0.05. Significance post hoc by
Dunn’s compared with health: *P<0.05, ** P<0.01.
healthy controls; mild asthma; moderate asthma; severe
asthma.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
118
Figure 3.14 Ratio of TH2:TH1 cells in different tissue compartments
Ratios of TH2 to TH1 cells in different tissue compartments, measured by intracellular cytokine staining
and flow cytometry. Differences are compared by Kruskal-Wallis tests.
healthy controls; mild asthma; moderate asthma;
severe asthma.
Peripheral TH2 responses correlate with atopy and with BAL TH2 cytokines
This evidence of elevated TH2 cell frequencies in peripheral blood implies that the TH2 bias in airway
tissues is part of a wider systemic TH2 bias. Woodruff et al used gene expression analysis of airway
epithelial cells to identify two subgroups of asthmatics which they termed ‘TH2-high’ and ‘TH2-low’ and
which differed according to biopsy expression of IL-5 and IL-13 as well as AHR, serum IgE and blood
and airway eosinophilia (Woodruff, Modrek et al. 2009). I therefore investigated whether similar
associations could be replicated in my data-set, and sought to stratify asthmatic subjects into TH2-high
and TH2-low subjects based on PBMC, sputum, BAL and biopsy TH2 frequencies. Defining TH2-high
as the top tertile of TH2 cell frequencies in each tissue type gave the greatest statistical power
(compared for instance with dichotomising at the median frequency), and also provided a good
differentiation from the normal range observed in healthy controls. For instance 1/3 of asthmatics had
a PBMC TH2 frequency ≥0.44%, whilst this was true for only 9% of healthy controls. I then tested the
variables identified by Woodruff using univariate analyses. Asthmatic subjects with high TH2
frequencies in peripheral blood had higher rates of atopy (100% v 77%, Fisher’s exact P=0.04),
responded to a greater range of allergens on skin prick allergy testing (P=0.002, Figure 3.15 A) and
also had higher BAL IL-5 levels (P=0.02, Figure 3.15 B).
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
119
Figure 3.15 Correlates of high peripheral blood TH2 frequencies
Asthmatic subjects stratified according to their peripheral blood frequencies of IL-13+ (TH2) cells as
TH2 high (the top tertile of PBMC TH2 frequencies) or TH2 low (lower two tertiles). (A) TH2 high
subjects tended to respond to more allergens on skin prick testing, t test P=0.002). (B) TH2 high
subjects also tended to have higher IL-5 levels in BAL (t test on Ln transformed data, P=0.02).
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
120
Figure 3.16 Compartmentalisation of tissue CD4+ T cells
Distribution of different CD4+ T cell subsets according to tissue type in all subjects combined. (A)
TH17 cells, (B) TH1 cells, (C) TH2 cells, (D) T reg. Groups are compared by Kruskal-Wallis tests with
post hoc Dunn’s. * P<0.05, ** P<0.01, *** P<0.001.
Similarly asthmatic subjects in the top tertile of sputum TH2 frequencies were more likely to be atopic
(P=0.04) and have more bronchial hyper-reactivity (P=0.02), whilst asthmatic subjects in the top tertile
of bronchial biopsy TH2 frequencies responded to a greater range of allergens on skin prick testing
(P<0.0005) (data not shown).
Distinct tissue localisation of different T cell subsets
It is apparent from figures 3.12 and 3.13 that different T cell subsets differ in their tissue distributions.
This compartmentalisation is analysed in detail in Figure 3.16. Both TH17 (A) and TH1cells (B) are
markedly concentrated in tissue compared with peripheral blood. Highest frequencies of TH17 cells
are observed in sputum and biopsies, whilst TH1 cells are most strongly localised to the
bronchoalveolar compartment. By contrast Treg were found at lowest frequency in biopsy tissue,
whilst I did not observe any significant tissue localisation of TH2 cells. These different tissue
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
121
localisations would be consistent with a dominant role of BAL effector / memory TH1 cells in immunity
to viruses (Cautivo, Bueno et al. 2010) or mycobacteria (Silver, Zukowski et al. 2003) whilst TH17 may
be more associated with mucosal immunity against bacterial or fungal invasion of stromal
tissue(Veldhoen and Stockinger 2006; Ma, Chew et al. 2008; Michel, Mendes-da-Cruz et al. 2008).
No evidence for a significant role of TCR+ IL-17+ T cells in human asthma
One cell type commonly considered to have a specific association with mucosal tissue is the innate-
like -T cell subset (Vanaudenaerde, Verleden et al. 2011). These T-cells are activated via their
TCRs and toll-like receptors (TLRs) and can provide a rapidly available source of IL-17. A
significant role for T-cells in allergic airways disease has been implied by animal models (Isogai,
Athiviraham et al. 2007; Jin, Roark et al. 2009). It has even been suggested that IL-17 secreting T-
cells may outnumber TH17 cells in murine allergic airway inflammation where they seem to be
critically involved in injury repair (Murdoch and Lloyd 2010). I therefore analysed IL-17 and IFN-
secreting T-cells in PBMC and BAL from a subset of 9 healthy controls and 24 asthmatics (Figure
3.17), using an antibody specific to all TCRs. According to my findings, T-cells are rare,
comprising only 1.9% (1.3-3.0%, median and IQR) of PBMC and 1.1% (0.65-2.8%) of BAL
lymphocytes, with much lower detectable frequencies of cytokine secreting cells. Therefore it was not
possible to enumerate T-cells in sputum or bronchial biopsy samples. Nonetheless there were no
significant differences in IL-17 secreting (A) or dual IL-17/IFN- secreting (B) T-cells in PBMC or
BAL. Although distributions of IFN- secreting BAL T-cells differed between groups (C), frequencies
were not different between health and any asthma phenotype. Thus my data do not provide evidence
of appreciable numbers of IL-17 secreting T-cells in humans, nor of any association with asthma
during periods of clinical stability. However these conclusions are limited by smaller sample sizes than
those used for other comparisons.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
122
Figure 3.17 γδ T cells in asthma
Frequencies of γδ T cells secreting (A) IL-17, (B) both IL-17 and IFN-γ and (C) IFN-γ as a proportion
of total γδ T cells in peripheral blood and in bronchoalveolar lavage. Groups are compared by
Kruskal-Wallis tests.
healthy controls; mild asthma; moderate asthma; severe
asthma. † median 0, IQR 0-0.05%.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
123
Figure 3.18 No evidence for TH2/17 cells in humans
A representative cytometry plot from subject 211 with moderate allergic asthma showing abundant
TH17 and TH2 cells, but no evidence of dual-cytokine secreting cells. Some PE bright cells spill into
the TH2/17 quadrant due to imperfect compensation.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
124
Figure 3.19 CD4+ T cell frequencies stratified by inflammatory cell subtype
(A) TH17 cells, (B) TH1 cells, (C) TH2 cells, (D) Treg cells according to tissue and inflammatory
subtype. Subjects have been classified according to sputum cell differentials as eosinophilic,
neutrophilic or paucicellular. A single individual had mixed eosinophilic / neutrophilic disease but was
classified as eosinophilic as this was the dominant feature. There were no statistically significant
differences between groups.
eosinophilic, neutrophilic, paucicellular.
No evidence for IL-17 producing TH2 cells in human asthma
One group recently reported the existence of IL-17-producing “TH2 cells” which express both the TH2
transcription factor GATA3 and the TH17 transcription factor RORt. Furthermore they observed
increased frequencies of these cells in peripheral blood of subjects with atopic asthma, and presented
murine data implicating them in the pathogenesis of experimental allergic airways disease (Wang,
Voo et al. 2010). In relation to these findings, I could not detect any evidence for dual IL-17/IL-13
secreting T cells in any tissue compartment. A representative plot is shown in Figure 3.18, where it
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
125
can be seen that despite high levels of IL-17 and IL-13 secretion, no dual secreting cells were
observed (beyond a minor compensation artefact). I did not however measure surface expression of
CCR6 or CRTH2, which were used by Wang et al to define the IL-17 secreting TH2 cells.
Analysis of CD4+ T cells according to inflammatory subtype
An important means to differentiate asthma into distinct endotypes is classification according to
inflammatory cell subtype (Simpson, Scott et al. 2006; Anderson 2008). Such fundamentally differing
patterns of airways inflammation suggest different underlying immunological processes. Therefore I
endeavoured to analyse my clinical and immunological data-sets in asthmatics according to their
differing inflammatory subtypes based on the sputum differential cell count. Inflammatory subtypes
were indeed related to clinical variables; specifically, logistic regression analysis showed that history
of allergic rhinitis was common (65% prevalence) in all subtypes except neutrophilic asthma (14%
prevalence, P<0.0001), implying that neutrophilic asthma is not strongly driven by allergic
inflammation of the nasal mucosa.
Other factors, whilst statistically significant by logistic regression, were likely to be artefacts of my
subject selection. Thus I observed a lower FEV1 in granulocytic asthma (mean 74.7% predicted FEV1
in eosinophilic, neutrophilic or mixed subtypes) compared with pauci-cellular asthma (mean 95.0%
predicted, P=0.0009) and a lower FEV1/FVC ratio in eosinophilic asthma (FEV1/FVC=62.6%)
compared with other subtypes (73.9%, P=0.0036). I also observed worse symptomatology in
neutrophilic asthma (ACQ 2.54 compared with 1.39 in other phenotypes, P=0.0007). However all of
these are expected consequences of my targeted recruitment of subjects with severe neutrophilic or
severe eosinophilic asthma from the Wessex Severe Asthma Cohort. A valid investigation of whether
these specific inflammatory subtypes are associated with worse lung function and symptom scores
would require an unbiased assessment of a much larger cohort of unselected asthmatic subjects.
Next I analysed CD4+ T cell frequencies according to inflammatory subtype. Data were available from
10 subjects with eosinophilic asthma, 15 with neutrophilic asthma and 25 with pauci-cellular asthma.
A single individual had mixed eosinophilic/neutrophilic disease but was classified as eosinophilic as
this was the dominant feature. There were no statistically significant differences between groups for
any major CD4+ T cell subset (Figure 3.19).
Cluster analysis to explore relationships between variables
With Dr Borislav Dimitrov I further explored possible associations, or clusters, among the various
(n=77) demographic, clinical and immunological parameters which described my cross-sectional
cohort using principle component analysis (PCA), as described in chapter 2. PCA has two purposes:
first to reduce the dimensionality (number of variables) of a high-dimensional data-set into a smaller
set of composite variables, much as multiple questions in an ACQ or quality of life score can be
reduced to a single summary statistic. The second purpose of PCA is to establish relationships
between variables and with the outcome of interest: in this case the presence and severity of asthma.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
126
Using an iterative process we reduced the 77 starting variables to a selected list of 25 variables,
which together were reduced by the PCA into 12 components, as shown in table 3.6.
Table 3.6 Principle component analysis of data from the cross sectional study.
1 2 3 4 5 6 7 8 9 10 11 12
Classification Mild/Mod/Sev
0.577
Age
Gender 0.879
Allergic Rhinitis 0.689
Eczema 0.926
ICS Dose 0.649
Nasal Polyps 0.808
Smoking History (PackYears)
0.892
ACQ 0.763
BMI 0.803
FEV1%Pred -0.789
GINA class 0.783
eNO 0.883
Total IgE 0.951
PBMC MAIT cells 0.837
PBMC TH17/Reg
Ratio 0.983
PTH1 0.917
PTH17 0.989
PTH2 0.645
Serum IFN-γ 0.512
Serum IL-10 0.987
Serum IL-13 0.987
Serum IL-17 0.607 -0.57
Serum IL-5 0.987
BAL MAIT cells 0.552 0.641
Component
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 21 iterations.
The table is a rotated component matrix. Each column represents a component, ranked from left to
right according to the extent to which they explain the variance in the data. Each cell gives a Pearson
correlation coefficient for the variable within the component. Only values >0.50 are shown.
What is the interpretation of this analysis? It should be viewed as hypothesis-generating because the
data were not all normally distributed and the sample size was small (n=31 asthmatic subjects) due to
missing data. However further analysis showed the components were robust, which means that the
analysis was significantly affected if components were removed. Together these 12 components
describe 92.6% of the variance within the dataset.
Component 1 is the most powerful component, explaining 15.7% of the variance alone. This
component shows that several T cell serum cytokine levels cluster together with each other and with
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
127
the dose of ICS, implying these variables are significantly correlated. Interestingly serum IL-17
clusters separately from these other cytokines in component 2, where it is significantly correlated with
three measures of asthma severity: ACQ score, GINA classification of disease control and the primary
classification I have used mild / moderate / severe. Although serum IL-17 was not different between
health and asthma overall, this result, and Figure 3.3 show that within asthma serum IL-17 levels are
higher in more severe disease. This is a different pattern from that observed with BAL fluids, is based
on a smaller sample and requires confirmation in a separate validation set. Other components
suggest the existence of distinct asthma endotypes such as the association of nasal polyps and
smoking in component 4 or the association of allergic rhinitis and blood TH1 cells in component 8. Of
relevance to chapter 5, MAIT cell frequencies in blood and BAL cluster with each other as
independent variables, and specifically do not cluster with ICS dose.
Discussion
The fundamental role of TH2 inflammation in asthma
My data provide the first comprehensive review of TH17 cells and Treg in the human airways, and set
them in the context of the well-characterised TH1 and TH2 cell subsets. With respect to the latter I
have observed significant increases in TH2 cells in both peripheral blood and airway tissues,
consistent with extensive prior literature (Robinson, Hamid et al. 1992; Anderson and Coyle 1994;
Cho, Stanciu et al. 2005; Woodruff, Modrek et al. 2009; Finkelman, Hogan et al. 2010; Lloyd and
Hessel 2010). Indeed my data add to the seminal findings of Robinson et al by extending the work to
a much wider spectrum of asthma. Robinson et al studied 15 mild, allergic, steroid naïve asthmatics
out of their allergen season, whilst my study has also included subjects on maintenance steroids and
subjects with severe asthma requiring high dose inhaled or oral steroids. It is apparent from both the
MSD and the cytometry data that the TH2 bias is less marked in these individuals. This may be due to
the effect of steroids in supressing TH2 responses or due to different underlying pathological
processes, or to a combination of the two. It is also apparent that this TH2 bias is a systemic
phenomenon, as it is observable in each tissue compartment tested, although the difference is most
marked in the lumen of the airway wall, as the greatest difference in median TH2 frequencies (a 12
fold difference between health and mild asthma) was observed in bronchial biopsies.
Evidence for a deficiency of regulatory T cells
In addition to the increased airway TH2 cells, I observed a deficiency of Treg in BAL in asthma which
was most pronounced in the most severe asthmatics. Treg share a reciprocal developmental
relationship with TH17 and have evolved to regulate tissue inflammation. The balance of Treg and
TH17 cell differentiation from naïve T-cells is regulated by TGFβ, IL-6, IL-21 (Bettelli, Carrier et al.
2006), Vitamin A and D, and the aryl hydrocarbon receptor (Quintana, Basso et al. 2008). However
the TH2 cytokine IL-4 can also influence Treg frequencies by blocking induction of FOXP3 Treg by
TGFβ (Dardalhon, Awasthi et al. 2008) and therefore a deficiency in Treg might be expected in a TH2
mediated disease such as asthma. In animal models Treg have been shown to suppress TH2
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
128
mediated allergic airway inflammation (Wu, Bi et al. 2008) and to mediate tolerance to chronic
aeroallergen exposure (Strickland, Stumbles et al. 2006).
Several authors have studied Treg in peripheral blood in human asthma. Mamessier et al studied 18
frequently-exacerbating severe asthmatics and observed lower frequencies of CD25hi Treg in blood
compared with 14 healthy controls (Mamessier, Nieves et al. 2008). They also observed a fall in the
frequencies and suppressive activity of peripheral Treg during exacerbations. Others have found
similar numbers of peripheral CD25Hi cells, but a fall in their FOXP3 expression in asthma (Lin, Shieh
et al. 2008; Provoost, Maes et al. 2009). Likewise Want et al found lower peripheral CD25Hi Treg in
asthma, correlating with higher allergen-induced IL-4 responses, whilst subjects who were atopic but
asymptomatic had higher levels of allergen-induced IL-10 implying a protective effect of Treg (Wang,
Lin et al. 2009). A key immunosuppressive mechanism for Treg is the production of IL-10 (Belkaid,
Piccirillo et al. 2002) which may be important for suppression of AHR (Kearley, Barker et al. 2005)
and a deficiency of IL-10 secreting cells in peripheral blood has also been reported in severe
compared with mild asthma(Matsumoto, Inoue et al. 2004; Hawrylowicz 2005). These associations
are not however straightforward as in paediatric populations others have found increases in IL-10 and
FOXP3 associated with presence of allergy (McLoughlin, Calatroni et al. 2012) or of more severe
asthma(Lee, Yu et al. 2007). Furthermore recently it has been shown that some FOXP3+ human
memory Treg can express RORt and secrete IL-17 but suppress effector T cells via cell-cell contact
(Ayyoub, Deknuydt et al. 2009; Voo, Wang et al. 2009).
To date few studies have examined Treg in the human airways. Heier et al demonstrated that some
FOXP3+ Treg were present within bronchus associated lymphoid tissue in infants with chronic
wheeze, but they did not investigate adults or healthy controls (Heier, Malmstrom et al. 2008). My
colleague Asha Ganesan investigated airway Treg in 10 mild asthmatics, 10 moderate asthmatics and
10 healthy controls. She observed a decrease in FOXP3+ Treg in sputum in mild-moderate asthma
(mean frequency 7.3% versus 11.8% in health, P=0.001)(Ganesan 2010). She also observed a
similar decrease in Treg in BAL, although it was significant only for moderate asthma (mean Treg
frequency 9.2% in health, 7.8% in mild asthma, 6.5% in moderate asthma, P=0.04) and found an
increase in the sputum TH17:Treg ratio in asthma.
Although I found no significant differences in peripheral blood Treg frequencies and did not replicate
Dr Ganesan’s findings in sputum, I did observe a similar deficiency of BAL Treg in asthma.
Furthermore, as with Dr Ganesan’s data, the difference correlated with disease severity (P=0.02),
being most marked in the most severe disease and was of similar magnitude (1.6-1.8 fold). Thus my
data constitute an important confirmation of her findings and have extended the observations to a
more severe phenotypic group.
Could this Treg deficiency be secondary to steroid treatment? Dr Ganesan’s moderate asthmatic
cohort were similar to my moderate cohort, with a median FEV1 of 98.1% (IQR 84.0-103) and
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
129
receiving a median equivalent of 400 mcg of beclometasone dipropionate per day, whist my severe
cohort were receiving 1600 mcg / day equivalent BDP (IQR 1280-2000). However a review of the
literature suggests that glucocorticosteroids actually tend to increase Treg frequencies in vivo in
murine peripheral lymphoid tissue (Chen, Oppenheim et al. 2006) and in human asthma in peripheral
blood where FOXP3 mRNA expression is increased by inhaled or oral steroids and correlates with IL-
10 mRNA expression (Karagiannidis, Akdis et al. 2004; Robinson, Larche et al. 2004; Provoost, Maes
et al. 2009). Whilst Seissler et al observed differing effects of steroids on different Treg subsets, they
too found that steroids induced the strongest increase in the subset of Tregs which were the most
suppressive (Seissler, Schmitt et al. 2012). Whilst each of these human studies sampled only
peripheral blood, it seems unlikely that the decrease in BAL Treg frequencies is due to steroids,
unless steroids were somehow reducing migration of Treg from blood to BAL.
Absolute frequencies are not the only relevant metric of Treg populations, as there is evidence that
steroids can increase the IL-10 production and suppressive activity of Treg(Robinson, Larche et al.
2004), or conversely that T cells from refractory asthmatics may be less able to produce IL-10 in
response to dexamethasone(Hawrylowicz, Richards et al. 2002). Therefore future studies need to
supplement measurement of Treg frequencies with functional assays of human airway Treg function
(Ganesan 2010), preferably before and after steroid treatment. Such work is likely to be worthwhile
because of the therapeutic potential of induction of allergen specific Treg by immunotherapy
(Robinson, Larche et al. 2004).
The uncertain significance of interleukin-17
With respect to interleukin-17 my findings were at odds with my prior hypotheses and with widespread
opinion, as an important role for IL-17 has been hypothesised by many authors in recent years
(Linden 2001; Aujla, Dubin et al. 2007; Anderson 2008; Alcorn, Crowe et al. 2010; Lloyd and Hessel
2010; Park and Lee 2010). However whilst the last decade has seen some excellent investigation of
IL-17 in animal models (Schnyder-Candrian, Togbe et al. 2006; Wakashin, Hirose et al. 2008; Wilson,
Whitehead et al. 2009; Lloyd and Hessel 2010; Murdoch and Lloyd 2010) the case for IL-17 in human
asthma has rested on just three papers (Molet, Hamid et al. 2001; Barczyk, Pierzchala et al. 2003;
Chakir, Shannon et al. 2003), which together have been cited over 150 times in the literature, and
which I must briefly address in the following:
In 2001 Molet et al provided the first description of IL-17 in the lungs of asthmatics using ELISA and
immunocytochemistry to show an increased number of IL-17+ cells in sputum and BAL from
asthmatics (Molet, Hamid et al. 2001). However this was a very small study, including only six
asthmatics, and it used an insensitive ELISA (LOD 5 pg/ml). More importantly, as with the other two
papers by Barczyk and Chakir, the ELISA technique used could not determine the cellular source of
IL-17. Indeed the only co-localisation data presented (by in situ hybridization) showed IL-17
production by eosinophils. In 2003 the same group reported further immunocytochemistry showing IL-
17+ cells were increased in the submucosa in moderate-to-severe asthma (Chakir, Shannon et al.
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
130
2003). Again this was a small study involving only six healthy controls and was unable to determine
the cellular source of the IL-17, though the localisation was to the submucosa. In the same year, the
above-mentioned paper by Barczyk and colleagues, which has since been widely cited paper, was
published (Barczyk, Pierzchala et al. 2003), proposing an association between sputum IL-17 levels
and BHR, which is clearly at odds with my data. Concerning the methodology of this study, it included
only 10 asthmatics, and the statistical report presented a post hoc analysis which may have resulted
from subgroup selection. Furthermore the ELISA technique used to measure IL-17 was most likely not
wholly appropriate, given a LOD of 15 pg/ml compounded by a reported 3-4 fold decrease in IL-17
levels in the presence of DTT. By contrast, the MSD assay used in my studies had a LOD of 0.22
pg/ml, a more modest effect of sample processing with DTE, and IL-17 levels were below 10 pg/ml in
all subjects of my study.
More recently Doe et al in Leicester suggested a slightly increased IL17+ submucosal staining in mild-
moderate (P=0.04) but, this time, not severe asthma (Doe, Bafadhel et al. 2010). This group used the
same IL-17 ELISA as Barczyk and for the reasons I have outlined it is not surprising that all 56
samples were below the limit of detection. The authors went on to measure IL-17 by MSD in 165
asthmatics, but unfortunately they included no healthy controls for comparison (ibid Fig 3). This paper
did however report more immunohistochemistry data suggesting a slight increase in IL-17+ cells in the
submucosa in mild-moderate asthma which would be consistent with Chakir et al, and with
unpublished immunohistochemistry data from Nivenka Jayasekera showing increased IL-17+ in the
epithelium of severe asthma (Jayasekera 2013). Jayasekera found that amongst these severe
asthmatics IL-17 correlated negatively with PEF and FEV1, possibly implying a protective role.
Furthermore whilst she found no correlation between staining for IL-17A and IL-17F or eNO levels,
she did observe an increase in IL-17A in mild asthma after allergen challenge.
How can these findings be brought together into a coherent concept? No group has yet produced
compelling data for a significant role for IL-17 in severe neutrophilic asthma, and my data would argue
strongly against such a role, at least in stable disease. There is no robust evidence of a relationship
between airway IL-17 and bronchial hyper-reactivity and my study provides strong data that there is
no such relationship. The work of Chakir, Doe and Jayasekera and others (Vazquez-Tello, Semlali et
al. 2010; Howarth 2012) do suggest a modest increase in expression of IL-17 in the cells of the airway
mucosa or submucosa in asthma. No agreement exists yet on whether this is predominantly IL-17A or
IL-17F (an issue complicated by the use of cross reactive antibodies in humans (Lloyd 2012)) or
whether IL-17 is pathogenic or protective (Murdoch and Lloyd 2010), and the exact cellular source of
this cytokine is similarly contentious. IL-17 can be produced by a wide variety of cell types including T-
cells, NK and NKT cells, tissue inducer lymphocytes, macrophages, and mast cells, eosinophils,
neutrophils, and epithelial cells (Molet, Hamid et al. 2001; Reynolds, Angkasekwinai et al. 2010;
Saitoh, Kusunoki et al. 2010). My data suggest that in stable asthma neither T-cells nor macrophages
constitute a major cellular source. Lack of a correlation with neutrophil numbers would also argue
against their being the primary source. It seems much more likely in the light of the
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
131
immunohistochemistry that the predominant cellular sources in the asthmatic human airway are either
eosinophils or bronchial epithelial cells.
Several investigators have demonstrated IL-17 production by airway eosinophils (Molet, Hamid et al.
2001; Saitoh, Kusunoki et al. 2010). In my data-set BAL IL-17 levels were correlated moderately with
BAL eosinophilia and eNO (Figure 3.7) and there was also a tendency towards an eosinophilic
sputum inflammatory subtype which was not statistically significant probably only because of the
small sample size (n=4)(Figure 3.10).
The case can also be made for epithelial cells as the dominant IL-17 producers. Epithelial cells are
numerically much more abundant than inflammatory cells of the airway and in my study high IL-17
secretion was strongly associated with unusually high shedding or sloughing of epithelial cells into the
BAL. Whilst initial research viewed the epithelial cells as downstream effectors in the IL-17 pathway
(Fossiez, Djossou et al. 1996; Linden 2001; Chen, Thai et al. 2003; Huang, Kao et al. 2007; Wiehler
and Proud 2007) it has recently been shown in mice (Suzuki, Kokubu et al. 2007; Ishigame, Kakuta et
al. 2009) and humans (Xu, Zhang et al. 2010) that airway epithelial cells also constitute a significant
source of IL-17F. Furthermore immunohistochemistry has shown epithelial staining for IL-17A (Chakir,
Shannon et al. 2003). Jayasekera interpreted this IL-17A staining as predominantly cytoplasmic
staining of epithelial cells, implying that it was unlikely to be cytokine bound to surface IL-17 receptors
(Jayasekera 2013). The possibility that human bronchial epithelial cells are producing IL-17A is a
hypothesis which warrants confirmation, for instance by PCR on pure epithelial cells.
Relegating TH17 cells
The lack of supporting evidence for pathological relevance of IL-17 in asthma in my studies of blood,
BAL, Sputum and tissue goes hand in hand with my central observation that TH17 cell frequencies do
not correlate with any phenotype of asthma. Hence, my findings, whilst confirming the centrality of the
TH2 response in asthma, relegate TH17 cells to a very minor role. Only few prior published data are
available to dispute it. One Chinese group reported differences in peripheral TH17 cell frequencies in
moderate-severe but not mild asthma, but their methods for statistical analysis were unclear, P values
not reported and there was a wide overlap between the group distributions (Shi, Shi et al. 2011). Most
importantly, airway tissues were not analysed in that study. Others have also looked in peripheral
blood but not tissue, such as Wong et al who measured peripheral IL-17 secreting cells by ELISA and
by surface markers(Wong, Lun et al. 2009). As TH17 cells are defined by their expression of IL-17 It is
not appropriate to enumerate TH17 cells by surface markers alone, and the marker used in this case –
CCR6 – is known to be expressed on both TH17 and FOXP3 Treg, as well as other T-cell subsets.
Bullens et al also claimed to have measured TH17 cells using a different method (Bullens, Truyen et
al. 2006). They report an increase in IL-17 mRNA in whole sputum which is moderately correlated
(r=0.5) with mRNA for CD3. Given the weakness of the association this is far from an accurate
method of enumerating TH17 cells, and the statistical analysis also raises concerns. Another paper
sometimes cited as evidence of TH17 in asthma is that by Pene et al who obtained T cell clones from
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
132
bronchial samples (Pene, Chevalier et al. 2008). Whilst this paper proved that TH17 cells are present
in the airway, it included only three subjects, all of whom were asthmatics sampled during an
exacerbation, and crucially they included no healthy controls. Finally Al-Ramli et al reported an
increase in submucosal IL-17+ cells using PCR and immunohistochemistry on bronchial biopsies.
However this paper did not present data on co-localisation so again there is no evidence that T cells
were the source of the IL-17 (Al-Ramli, Prefontaine et al. 2009).
Others have reported a subset of TH2 cells, defined as expressing the surface marker
chemoattractant receptor-homologous molecule expressed on TH2 cells (CRTH2), which also
expressed IL-17 (Cosmi, Annunziato et al. 2000; Wang, Voo et al. 2010). These cells were found at
higher frequencies in peripheral blood of 23 atopic asthmatics(Wang, Voo et al. 2010). While I did not
stain for CRTH2, I have been unable to detect any dual IL-13/IL-17 T cells in my dataset, although
this may partly be because IL-13 expression was not maximal by 4-5 hours. Cosmi et al reported T
cell clones which express both IL-17 and IL-4, and found them to be more frequent in asthma (n=11),
although they acknowledge these cells are extremely rare, comprising approximately 0.025% of T
helper cells in health, and like me they found no difference in TH17 frequencies between asthma and
health(Cosmi, Maggi et al. 2010).
In summary, the data presented in this chapter provide strong evidence that TH17 cells are not
associated with asthma in stable disease. The study subjects did not undergo allergen challenge,
although they were sampled throughout the year, so many will have received ongoing exposure to
perennial and seasonal allergens. Finally this cohort was not sampled within six weeks of a
symptomatic viral illness. It is possible that TH17 cells might play a more important role during an
acute antiviral immune response, and so I present data on T cell responses during natural
exacerbations in chapter 7.
T-cells
I did not find evidence of high frequencies of 17-secreting T-cells associated with asthma. This is
perhaps an instance where there is a distinct species difference between the immunology of mice and
humans. In mice T-cells have been implicated in the pathogenesis of experimental allergic airways
disease, being necessary for the IL-4 dependent generation of specific IgE and IgG1, of pulmonary IL-
5 and -13 and in recruiting T cells and eosinophils to the airways (Zuany-Amorim, Ruffie et al. 1998;
Jin, Roark et al. 2009). Other data suggest they may also play a subsequent role in the resolution of
airways inflammation as CD8+ T-cells (in rats) (Isogai, Athiviraham et al. 2007) or IL-17 secreting
T-cells (in mice) (Murdoch and Lloyd 2010) can decrease AHR, the late allergic airway response,
eosinophilia and TH2 responses. In mice the dominant IL-17-producing cells in the spleen are T-
cells more than TH17 cells(Stark, Huo et al. 2005) and likewise in a murine OVA challenge model the
dominant IL-17 secreting cells in BAL were -17 cells more than TH17 cells (Murdoch and Lloyd
2010). Conversely in humans I have observed much greater numbers of TH17 cells than -17 cells in
each tissue compartment. In BAL T-cells are rare, comprising only 1.1% of lymphocytes, of which a
Timothy SC Hinks 3. CD4+ T cell phenotypes in asthma
133
median 5.6% secreted IL-17, whilst the majority (42.3%) secreted IFN-. Perhaps significant species
differences in T-cells biology are unsurprising as similar differences between mouse and man have
been observed with other innate-like lymphocytes, namely the iNKT cells which are found at higher
frequencies in mice and MAIT cells which, conversely, are 5-10 fold more abundant in humans
(Treiner, Duban et al. 2005). MAIT cells will be the subject of the next chapter.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
134
CHAPTER 4
CD8+ T cells in asthma The most incomprehensible thing about the world
is that it is comprehensible.4
4 Albert Einstein (1879-1955), cited in Vallentin, A. (1954). Einstein: A Biography. London,
Weidenfeld and Nicolson. p24
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
135
Introduction
In order to place T helper subsets in context I also undertook a parallel analysis of CD8+ ‘cytotoxic’ T
cells (TCYT). These cells have received much less attention in airways disease, although a few studies
have shown that airways disease in COPD may be associated with an increase in epithelial (Fournier,
Lebargy et al. 1989) or subepithelial (O'Shaughnessy, Ansari et al. 1997; Saetta, Di Stefano et al.
1998) CD8+ T cells. In asthma CD8+ T cells have been found in increased frequencies and activation
state in post-mortem specimens (O'Sullivan, Cormican et al. 2001) and one longitudinal study has
shown a modest correlation between bronchial biopsy CD8+ T cells and subsequent rate of decline in
lung function (van Rensen, Sont et al. 2005). Furthermore as a primary role of CD8+ T cells is direct
antiviral activity and in view of the increasing appreciation of the relevance of respiratory viral
infections in asthma, it seems timely to investigate these cells afresh (Johnston, Pattemore et al.
1995; Johnston, Pattemore et al. 1996; Corne, Marshall et al. 2002; Message, Laza-Stanca et al.
2008). As mentioned in chapter 1, CD8+ T cells form functionally distinct subsets known as Tc1 and
Tc2 cells, according to their expression of type 1 or type 2 cytokines respectively (Mosmann, Li et al.
1997). One previous studies have found evidence of an increase in the Tc2 subset in asthma in
sputum (Cho, Stanciu et al. 2005), and it is this cell type which will be the focus of this chapter.
In this chapter I will present cross-sectional data showing that the Tc2 subset of cells are increased in
asthma and are associated specifically with an eosinophilic endotype.
Results and comments
Study population
This analysis was performed on the samples taken from the same population which was described in
chapter 3, comprising 23 healthy subjects and 53 asthmatics (14 mild, steroid-naïve, 17 moderate,
treated with low dose inhaled corticosteroids and 22 severe, treated with oral or high dose inhaled
corticosteroids) were studied. All had stable symptoms for at least 6 weeks prior to clinical sampling.
Clinical and demographic characteristics are presented in Table 3.1.
Definitions of T cell subsets
T cells were stimulated ex vivo for 4 to 5 hours with PMA and ionomycin and analysed by flow
cytometry. Tc1 cells were defined as live CD3+CD8+ T cells expressing IFNγ. Tc2 cells were defined
as live CD3+CD8+ T cells expressing IL-13. Frequencies are expressed as a percentage of the total
CD8+ T cell population. Analysis using absolute numbers of Tc1 and Tc2 cells, expressed as a
proportion of CD3+ T cells, yielded the same findings.
Type 2 cytokine-secreting cytotoxic T cell frequencies are increased in asthma in PBMC and
BAL, and correlate with disease severity
I observed significant increases in the relative frequencies of CD8+ T cells secreting IL-13 (Tc2) in
asthma compared with health in both peripheral blood (n=66, Mann-Whitney P=0.04) and BAL (n=60,
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
136
P=0.02) with a similar pattern in bronchial biopsies, although the latter did not reach significance,
possibly due to the smaller sample size (n=48)(Figure 4.1).
Figure 4.1 Type 2 cytokine-secreting cytotoxic T cell frequencies are increased in asthma in
PBMC and BAL
Frequencies of CD8+ T cells which express the type 2 cytokine IL-13 (Tc2 cells) expressed as a
proportion of total live CD8+ T cells in peripheral blood, sputum, BAL and biopsies in health and
asthma. Box and whisker plots show medians and IQRs. Differences are compared by unpaired
Mann-Whitney tests and shown if P>0.05.
healthy controls n=19 PBMC, 12 sputum, 17 BAL, 13 biopsies.
asthma n=47 PBMC, 26 sputum, 43 BAL, 35 biopsies.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
137
Figure 4.2 Type 1 cytokine-secreting cytotoxic T cell frequencies are increased in BAL in
asthma
Frequencies of CD8+ T cells which express the type 1 cytokine IFN-γ (Tc1 cells) expressed as a
proportion of total live CD8+ T cells in peripheral blood, sputum, BAL and biopsies in health and
asthma. Box and whisker plots show medians and IQRs. Differences are compared by unpaired
Mann-Whitney tests and shown if P>0.05.
healthy controls n=19 PBMC, 12 sputum, 17 BAL, 13 biopsies.
asthma n=47 PBMC, 26 sputum, 43 BAL, 35 biopsies.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
138
Figure 4.3 Type 1 cytokine-secreting cytotoxic T cells are increase in BAL in mild asthma
Frequencies of CD8+ T cells which express the type 1 cytokine IFN-γ (Tc1 cells) expressed as a
proportion of total live CD8+ T cells in peripheral blood, sputum, BAL and biopsies in health and
asthma stratified according to disease severity. Box and whisker plots show medians and IQRs.
Differences are compared by Kruskal-Wallis tests with post hoc Dunn’s and are significant only for
health v mild asthma in BAL.
healthy controls n=19 PBMC, 12 sputum, 17 BAL, 13 biopsies.
mild asthma n=14 PBMC, 9 sputum, 13 BAL, 13 biopsies.
moderate asthma n=14 PBMC, 8 sputum, 14 BAL, 13 biopsies.
severe asthma n=19 PBMC, 9 sputum, 16 BAL, 10 biopsies.
Type I cytokine-secreting cytotoxic T cell are increased only in BAL, in mild asthma.
IFN-γ-secreting T cells (Tc1) did not differ significantly between asthma and health in PBMC, sputum
or bronchial biopsies, but were increased in asthma in BAL with a median frequency of 73% (IQR 80-
89%) compared with 84% (42-88%) in health (P=0.02) (Figure 4.2). When subjects were stratified
according to disease severity this difference is seen to result from increased frequencies of Tc1 cells
in mild asthma (median frequencies in mild asthma 88% (84-90%) compared with health 73% (42-
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
139
92%), Dunn’s P<0.01, Figure 4.3). This asthma-related increase in IFN-γ-secreting T cells is specific
to the CD8+ subset (Tc1) and is not observed in the CD4+ subset (TH1)(Figure 4.4).
Figure 4.4 A comparison of Tc1 and TH1 cells in BAL
Frequencies of (A) CD8+ Tc1 cells and (B) CD4+ TH1 expressed as a proportion of total live CD8+
and CD4+ T cells respectively in BAL in health and asthma stratified according to disease severity.
Box and whisker plots show medians and IQRs. Differences are compared by Kruskal-Wallis tests
with post hoc Dunn’s and are significant only for Tc1 in health v mild asthma in BAL. By contrast there
are no differences in TH1 cell frequencies in BAL.
healthy controls n=17
mild asthma n=13
moderate asthma n=14
severe asthma n=16
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
140
Figure 4.5 Frequencies of IL-17-secreting CD8+ T cell do not differ asthma
Frequencies of CD8+ T cells which express IL-17 (Tc17 cells) expressed as a proportion of total live
CD8+ T cells in peripheral blood, sputum, BAL and biopsies in health and asthma. Box and whisker
plots show medians and IQRs. No significant differences were observed between asthma and health
(Mann-Whitney test P>0.2 in all tissues, not shown) or between health and different phenotypes
(Kruskal-Wallis P>0.5 in all tissues) healthy controls.
healthy controls n=19 PBMC, 12 sputum, 17 BAL, 13 biopsies.
mild asthma n=14 PBMC, 9 sputum, 13 BAL, 13 biopsies.
moderate asthma n=14 PBMC, 8 sputum, 14 BAL, 13 biopsies.
severe asthma n=19 PBMC, 9 sputum, 16 BAL, 10 biopsies.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
141
Figure 4.6 Correlations between Tc2 and TH2 cells in tissues
Correlations between frequencies of TH2 and Tc2 cells in (A) PBMC, (B) BAL, (C), sputum and (D)
bronchial biopsies. Spearman’s correlations are presented.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
142
Figure 4.7 Type 2 cytokine-secreting cytotoxic T cell frequencies correlate with disease
severity in blood
Frequencies of CD8+ T cells which express the type 2 cytokine IL-13 (Tc2 cells) expressed as a
proportion of total live CD8+ T cells in peripheral blood, sputum, BAL and biopsies in health and
asthma stratified according to disease severity. Box and whisker plots show medians and IQRs.
Linear trends are compared across groups using Jonckhere-Terpstra test and are significant only for
PBMC.
healthy controls n=19 PBMC, 12 sputum, 17 BAL, 13 biopsies.
mild asthma n=14 PBMC, 9 sputum, 13 BAL, 13 biopsies.
moderate asthma n=14 PBMC, 8 sputum, 14 BAL, 13 biopsies.
severe asthma n=19 PBMC, 9 sputum, 16 BAL, 10 biopsies.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
143
Figure 4.8 Type 2 cytokine-secreting cytotoxic T cell frequencies according to inflammatory
subtype, nasal polyposis and history of smoking
(A) Frequencies of Tc2 cells in bronchial biopsies correlate strongly with the presence of eosinophilic
asthma (based on sputum inflammatory cell type)(Kruskal-Wallis P=0.03 across eosinophilic,
neutrophilic and paucicellular subgroups, post hoc Mann-Whitney P=0.006). (B) Biopsy TH2 cell
frequencies are also associated with eosinophilic asthma, but the association is less strong (Mann-
Whitney P=0.02). In peripheral blood Tc2 cell frequencies are associated with (C) a history of nasal
polyposis (Mann-Whitney P=0.008) and a history of ever smoking (Mann-Whitney P=0.008). Box and
whisker plots show medians and IQRs.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
144
Frequencies of IL-17 secreting cytotoxic T cells are not associated with asthma
I observed no differences in frequencies of IL-17-secreting CD8+ T cells between asthma and health
(P>0.2 in all tissues) or between health and any category of disease severity (P>0.5 for all tissues,
Figure 4.5).
Clinical correlations with Tc2 cell frequencies
Tc2 cell frequencies correlated with Th2 cell frequencies in BAL (rs=0.489, P=0.0001, n=57), and
weakly in sputum (rs=0.362, P=0.03, n=36), Tc2 and Th2 frequencies were not significantly correlated
in blood or biopsies Figure 4.6).
Peripheral blood Tc2 frequencies were higher in more severe disease (Jonckhere-Terpstra test,
P=0.01), with a similar pattern in BAL (ns)(Figure 4.7).
Type 2 cytokine-secreting cytotoxic T cell frequencies according to inflammatory subtype,
nasal polyposis and history of smoking
I observed a wide range of Tc2 frequencies, particularly in biopsies, suggesting that these patterns
may result from a large increase of Tc2 cells in a specific subset of individuals. I therefore explored
relationships with clinical characteristics in univariate analyses. Frequencies of bronchial biopsy Tc2
cells differed significantly between different inflammatory subtypes (Kruskal-Wallis P=0.03). This
difference was due to a striking 20 fold increase in Tc2 cells in eosinophilic asthma (median 2.1%,
IQR 0.53-2.5%) compared with other subtypes (0.10%, 0.0-0.48%, Mann-Whitney P=0.006)(Figure
4.8 A). For comparison in my data-set this difference is of greater magnitude than the much better
documented phenomenon of increased biopsy CD4+ TH2 cells in eosinophilic asthma (eosinophilic
asthma 1.7%, 0.4-3.3% versus other subtypes 0.3%, 0.1-1.1%, P=0.02)(Figure 4.8 B). This
phenomenon can also be observed as a much lower ratio of bronchial biopsy Tc1 to Tc2 cells in
eosinophilic asthma (median 27% (15-43%) than paucigranulocytic asthma (median 117% (86-472%,
P<0.05, Figure 4.9).
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
145
Figure 4.9 Ratio of TC1:TC2 T cells according to inflammatory subtype
The ratio of Tc1 to Tc2 cells in peripheral blood, sputum, BAL and biopsies in asthma stratified
according to sputum inflammatory cell subtype. Box and whisker plots show medians and IQRs. A
single subject with mixed subtype was classified as eosinophilic for this analysis as this was the
predominant feature. Differences are compared by Kruskal-Wallis tests (P values given) with post hoc
Dunn’s (denoted with * for P<0.05) and are significant in BAL and bronchial biopsies.
Eosinophilic n=9 PBMC, 3 sputum, 6 BAL, 5 biopsies.
Neutrophilic n=12 PBMC, 6 sputum, 12 BAL, 9 biopsies.
Paucigranulocytic n=14 PBMC, 10 sputum, 10 BAL, 6 biopsies.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
146
Figure 4.10 Clinical correlates of peripheral blood Tc2 cell frequencies
Frequencies of IL-13-secreting CD8+ T cells as a percentage of total CD8 cells are correlated
positively with (A) duration of asthma in years, (C) age in years and (D) BAL neutrophilia and
correlated negatively with FEV1% predicted. Spearman’s correlations are presented.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
147
Clinical correlates of peripheral blood Tc2 cell frequencies
Univariate analyses of peripheral blood Tc2 frequencies, showed higher frequencies were associated
with a history of nasal polyposis (P=0.008) and a history of ever smoking (P=0.008, Figure 4.8 C,D).
Using univariate Spearman’s correlations frequencies of Tc2 cells were correlated positively with
duration of asthma (rs=0.436, P=0.002, Figure 4.10 A), with age (rs=0.477, P=0.001, Figure 4.10 C)
and more weakly with BAL neutrophil count (rs=0.339, P=0.008, Figure 4.10 D). Finally Tc2
frequencies were correlated negatively with FEV1 (% predicated).
Preliminary analysis of the T cell transciptome is supportive of a role for CD8+ T cells in
asthma
In the discussion chapter I will outline my plans for a comprehensive analysis of the T cell the
transcriptome in pure populations of sorted CD3 T cells from blood and airway tissues. Numbers of
samples which were successfully hybridised are presented in table 4.1. Such comprehensive analysis
has not yet been performed, but I will briefly present the most preliminary analysis of this data-set, as
it is relevant to this chapter.
Table 4.1 Numbers of successful microarrays performed and passing quality data quality
control
Tissue Healthy
control
Mild asthma Moderate
asthma
Severe
Asthma
Total
PBMC 12 9 11 10 42
BAL 14 14 10 8 46
Sputum 5 7 4 8 24
Sputum post
ICS
6 6 N/A N/A 12
Epithelial
cells
12 12 10 8 42
Figure 4.11 shows a network analysis of peripheral blood CD3+ T cells revealing severe asthma is
associated with a significant down-regulation in T cell associated networks. Furthermore these genes
were not found to correlate with ICS dose. These data provide further evidence of the importance of T
cells in the pathogenesis of asthma, and imply that such responses are steroid resistant.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
148
Figure 4.11 T cell associated networks are down-regulated in severe asthma
Transcriptomic data from sorted peripheral blood CD3+ cells reveal that severe asthma is associated
with a significant down-regulation in genes associated with T cell networks. These genes did not
correlate with ICS dose.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
149
Figure 4.12 Hierarchical clustering of asthma v health in BAL T cells reveals a strong asthma-
associated gene signature
Hierarchical clustering was performed on gene lists from asthma v healthy with greater than ±1.5 fold,
P≤0.05 difference in gene expression (875 probe set IDs). Only male subjects were included because
of a strong gender effect.
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
150
Figure 4.13 Hierarchical clustering of asthma v health in sputum T cells reveals a strong
asthma-associated gene signature
Hierarchical clustering was performed on gene lists from asthma v healthy with greater than ±1.5 fold,
P≤0.05 difference in gene expression. Only male subjects were included because of a strong gender
effect.
Hierarchical clustering of genes differentially expressed (±1.5 fold, P≤0.05) between asthma and
health in BAL Figure 4.12 (875 genes) and sputum Figure 4.13 (1181 genes). Further comparisons
are given in Figure 4.13 (health versus mild asthma, health versus moderate asthma and health
versus severe asthma in sputum). The highest upregulated gene was CCL18 which is a chemotactic
for T cells and expressed at high levels in the lung. Pathway analysis identified ‘cytotoxic T
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
151
lymphocyte mediated apoptosis of target cells’ as the 5th most differentially expressed pathway in
asthma, providing further evidence of the importance of CD8+ T cells in asthma.
Thus this very brief report of T cell transcriptomics demonstrates that asthma is associated with
significant changes in the function of T cell networks, including cytotoxic T cells, which will merit
detailed investigation in the future through data-sets such as this (see chapter 8).
Discussion
In this brief cross-sectional investigation of CD8+ T cells I have observed a specific increase in the
Tc2 cell subset in asthma in PBMC and BAL, which is associated with increasing disease severity and
which is most striking in a subset of eosinophilic asthmatics. This subset were further characterised
as subjects who tended to have a history of smoking, of nasal polyposis, and older age.
What is known of a link between CD8+ cells and eosinophils in asthma?
Perhaps one reason that CD8+ cells have received so little attention is that several early studies have
using allergen challenge in humans found no increase in airway CD8+ T cells after allergen challenge
(Aalbers, Kauffman et al. 1993; Aalbers, Kauffman et al. 1993; Bentley, Meng et al. 1993) but instead
reported significant increases in airway eosinophils in BAL and bronchial biopsies. However these
studies did not compare asthmatics with healthy controls and focussed on different asthma
endotypes. Furthermore more recent studies have analysed not just absolute numbers of T cells, but
also their activation status and cytokine expression and this has led to different conclusions. Walker et
al. focussed on eosinophilic lung diseases, including allergic asthma and did find significantly
increased numbers of activated CD4+ and CD8+ T cells in BAL compared with health, observing a
close correlation between numbers of activated T cells, eosinophils and IL-5 levels (Walker, Bauer et
al. 1994). Using immunohistochemistry Ying et al. found that biopsy CD8+ cells as well as CD4+ cells
expressed IL-4 and -5 in asthma (Ying, Humbert et al. 1997).
Increases in CD8+ T cells have also been observed in other forms of asthma. Frew et al. compared
atopic asthma with red cedar asthma, and found the latter was characterised by a 4-fold greater
increase in biopsy T cells and 2.5 times greater increase in biopsy eosinophils than atopic asthma
(Frew, Chan et al. 1995). They specifically found increases in biopsy CD8+ T cells in a subset of
individuals with red cedar asthma. Another subtype of asthma is induced by toluene diisocyanate
(TDI) (a chemical intermediate in the production of polyurethane) and Finotto et al. observed that TDI
challenge in sensitised individuals induced a 56% increases in CD8+ T cells at 8 hours, followed at 24
hours by a 2.5 fold increase in eosinophils (Finotto, Fabbri et al. 1991).
Together these studies suggest a close relationship between CD8+ T cells and eosinophils. It is likely
that this is mediated, as least in part, by IL-4 and IL-5. Till et al. found asthma was associated with
increased IL-5 production in CD8+ T cell lines derived from BAL (Till, Li et al. 1995). Cho et al.
recently reported increased production of IL-4 and IL-5 by unstimulated sputum CD4+ and CD8+ cells
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
152
in asthma, which was more closely related to disease severity in CD8+ than CD4+ T cells (Cho,
Stanciu et al. 2005).
Data from animal models has given further insight into the mechanisms underlying these
observations, and suggests that viruses play an important role in the link between eosinophils and
CD8+ T cells. Allergen challenge in guinea pigs caused an increase in mucosal T cells which were
almost entirely CD8+ and were strongly correlated with eosinophils at 6 hours (Frew, Moqbel et al.
1990), but CD8+ cells are not pathogenic in all animal models (Huang, MacAry et al. 1999; Isogai,
Athiviraham et al. 2007; Jiang, Wang et al. 2009). These apparent inconsistencies seem to depend on
the interaction between viruses and allergen. In a mouse model of allergic airways disease bystander
allergen specific TH2 responses, mediated by IL-4, could re-programme virus-specific CD8+ T cells to
produce IL-5 and recruit eosinophils to the airways (Coyle, Erard et al. 1995). Furthermore when the
mice were re-challenged with virus specific peptides they responded by IL-5 production recruiting
further eosinophils to the airways. Likewise mice infected with respiratory syncytial virus (RSV)
developed lung eosinophilia and AHR which could be prevented by prior depletion of CD8+ cells
(Huang, MacAry et al. 1999; Schwarze, Cieslewicz et al. 1999). Similarly CD8+ cells were essential
for induction of eosinophil degranulation and AHR in guinea pigs infected with parainfluenza
(Adamko, Fryer et al. 2003). In this model it was allergen sensitisation which increased the number of
eosinophils in close relation to airway nerves, but the CD8+ cells mediated the release of major basic
protein by degranulation of eosinophils which induced AHR by blocking M2 muscarinic receptors.
Enomoto has shown that perforin (and thus cytotoxicity) is necessary for allergen-specific CD8+ cells
to modulate allergic inflammation (Enomoto, Hyde et al. 2012). Finally Sawicka et al. used adoptive
transfer experiments in mice to show it was the Tc2 cells, not Tc1 cells, which induced the
eosinophilia and AHR (Sawicka, Noble et al. 2004).
Together these animal data suggest a synergy between allergens and viruses leading to eosinophilic
AHR. This would explain the strong association in my data-set between increased numbers of Tc2
cells and eosinophils in biopsies. Indeed there is some evidence of such synergy occurring in
humans. Calhoun et al. performed segmental allergen challenge before and after experimental
infection with Rhinovirus type 16 (RV16) in 7 patients with allergic rhinitis and 5 healthy controls
(Calhoun, Dick et al. 1994). They found that allergen challenge induced greater histamine release and
eosinophil recruitment associated with RV16 infection and persisting for a month afterwards. This may
explain why inhaled corticosteroids, which potently decrease airway eosinophils, also reduce the
frequency of exacerbations in persistent asthma (Kelly and Busse 2008).
I also observed an association between peripheral blood Tc2 cells and BAL neutrophilia. Neutrophils
are the predominant inflammatory cell during exacerbations (Dougherty and Fahy 2009) and perhaps
the neutrophils I observed in subjects with Tc2 inflammation were a residual effect of previous
exacerbations, although none of my subjects had viral symptoms within the preceding 6 weeks and
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
153
viruses were not detectable in the BAL at the time of sampling. It would be informative to study the
time course of Tc2 responses during viral infection and correlate these with airway inflammatory cells.
The role of Tc1 cells in asthma
I have also presented data on other CD8+ T cells. Whilst I did not observe significant differences in IL-
17 secreting CD8+ T cells in any tissue, I did observe an increase in Tc1 cell frequencies in asthma.
This difference was only observed in the BAL tissue compartment, and was significant only in mild
asthma. Moreover this difference in IFN-γ secreting T cells was observed only amongst CD8+ cells
and not in the CD4+ T helper subset.
These observations may help reconcile apparently conflicting data in previous literature. Whilst much
type 2 cytokine secreting T cells have received much more attention in asthma, there are a handful of
previous reports regarding IFN-γ secretion in BAL cells in human asthma. Increased spontaneous
release of IFN-γ has been observed from BAL leukocytes in BAL (Cembrzynska-Nowak, Szklarz et al.
1993). Krug et al. further characterised these cells, observing increased IFN-γ secreting T cells in the
BAL from 10 asthmatic subjects (Krug, Madden et al. 1996). These subjects all had mild, steroid
naïve atopic asthma with a mean FEV1 of 100.4% predicted. Krug et al. subsequently reported that
allergen challenge caused a decrease in the proportion of BAL T cells secreting IFN-γ (Krug,
Erpenbeck et al. 2001). These findings were not in accord with the findings of Del Prete et al. who
reported that the majority of CD4+ T cell clones produced only low quantities of IFN-γ (Del Prete,
Maggi et al. 1988). Crucially, however, the papers by Krug et al. did not co-stain for CD4 and CD8,
and so were unable to distinguish between cytokines secreted by Tc1 and TH1 cells. My data would
suggest that the source of the increased IFN-γ in these studies of human asthma was the Tc1 cell
population.
The effect of IFN-γ in human asthma is unknown. It has been suggested, based on animal data, that
Tc1 cells may be capable of moderating inflammation and suppressing AHR (Betts and Kemeny
2009). In mice IFN-γ can inhibit airway eosinophilia and production of mucus and chitinases, as well
as inhibiting eosinophil production in the bone marrow (Sawicka, Noble et al. 2004; Mitchell, Provost
et al. 2011). Others have shown that IFN-γ can be induced 60 fold in CD4+ and CD8+ T cells by
factors produced by mast cell lines (de Pater-Huijsen, de Riemer et al. 2002). IFN-γ may have other
effects which are less beneficial in airways inflammation. Experiments in which allergen-specific TH1,
TH2 cells or both were adoptively transferred in mice showed that TH1 cells alone could induce airway
inflammation and lymphocyte recruitment and further that TH1 cells could facilitate TH2 cell
recruitment, to synergistically induce a more vigorous, eosinophilic inflammatory response (Randolph,
Stephens et al. 1999). This paper did not examine CD8+ cells, but one group have reported on Tc1
cells in a murine model. Allergen challenge of mice injected with Tc1 cells induced neutrophilic airway
inflammation but no induction of AHR, which was in contrast to the eosinophilia and AHR induced by
Tc2 cells (Sawicka, Noble et al. 2004). In contrast to this I did not observe an association between
BAL Tc1 frequencies and neutrophilic inflammation (Figure 4.9 C).
Timothy SC Hinks 4. Cytotoxic CD8+ T cells in asthma
154
Therefore my data suggest that asthma, particularly mild asthma, is associated with an increase in
BAL Tc1 cells which it is likely may have mixed effects on airway inflammation, though these are likely
to include inhibition of cardinal features of TH2 inflammation including airway eosinophilia, mucus
hyper-secretion and AHR.
Conclusion
Compared with T helper cells, CD8+ T cells have been the subject of little attention from asthma
researchers in recent years. My data suggest they may well play an important role in asthma,
particularly in a subset of individuals with eosinophilic inflammation of the airway mucosa. These
subjects tend to have distinct clinical features – namely a history of smoking, of nasal polyposis and
are older – suggesting they may represent a distinct clinical phenotype of asthma. There is likely to be
an important interaction between virus and allergen-induced inflammation with animal data suggesting
allergen-specific IL-4-mediated TH2 responses may re-programme virus-specific IL-5 mediated Tc2
cells to induce pathogenic airway eosinophilia and AHR. The role of Tc2 cells merits further work,
which should investigate the specificity of these cells, for example using tetramers, ELISpot, or
ELISpot arrays and should delineate their role in viral infections in asthma.
Timothy SC Hinks 5. MAIT cells – new players in asthma
155
CHAPTER 5
MAIT cells – new players in asthma The presence of bronchial asthma is much
more easily ascertained than is the cause.5
5 Rubin, E.H. and M. Rubin, Diseases of the Chest 1947, Philadelphia and London: W.B.
Saunders
Timothy SC Hinks 5. MAIT cells – new players in asthma
156
Introduction
Having investigated the roles of both the major adaptive T cell subsets – CD4+ T helper cells and
CD8+ cytotoxic T cells – I wish now to turn to the role of an emerging class of innate-like T cells: the
mucosal associated invariant T (MAIT) cell. The discovery in the last decade of innate-like T cells
which are restricted by MHC-like molecules and can respond rapidly to non-peptide antigens has
been an exciting development in our understanding of T cell biology (Kronenberg and Kinjo 2009), but
the role of one such cell – the iNKT cell - in the human airways has been an issue of some
controversy (Umetsu and Dekruyff 2006; Vijayanand, Seumois et al. 2007; Djukanovic and Gadola
2008; Meyer, DeKruyff et al. 2008). The more recent discovery of MAIT cells is of great interest. MAIT
cells share with iNKT cells the distinction of expressing conserved αβTCRs and being restricted by a
MHC class I-like nonpolymorphic molecules, i.e. MR1 (Treiner and Lantz 2006). MAIT cells are
relatively frequent in human peripheral blood, and they are five to tenfold more frequent in humans
than in mice, while iNKT cells seem to be more abundant in mice (Treiner, Duban et al. 2005). MAIT
cells are specifically activated by molecule MR1 which is the most highly conserved MCH class 1
related molecule in mammals, implying a functional role of key evolutionary importance (Brossay,
Chioda et al. 1998). Yet to date no data have been published on MAIT cells in relation to human lung
disease.
In this chapter I will describe the first analysis of MAIT cells within the human airways, in which I have
made the novel finding of a deficiency of MAIT cells in asthma. I will show how this deficiency
correlates with various clinical characteristics. I will then present exploratory research into the biology
of these cells including a characterisation of the cytokine expression profile of MAIT cell clones and
evidence of their potential to be modulated by corticosteroids and perhaps influenced by systemic
levels of vitamin D3.
Results and comments
Study population
Twenty healthy subjects and 54 asthmatics (15 mild, steroid-naïve, 22 moderate, treated with low
dose inhaled corticosteroids (ICS) and 17 severe, treated with oral or high dose inhaled
corticosteroids) were studied. All had stable symptoms for at least 6 weeks prior to clinical sampling.
The study design is shown in Figure 2.1 and clinical characteristics of the study participants are
shown in table 5.1.
Timothy SC Hinks 5. MAIT cells – new players in asthma
157
Table 5.1 Clinical characteristics of MAIT cell study population
n 20 15 22 17DemographicsSex (M/F) 12 / 8 8 / 7 9 / 13 7 / 10Age (median [range], years) 28 (24-43) 26 (22-33) 36 (24-47) 53 (42-63)Pulmonary function
FEV1 (% predicted) 107 (95-113) 88 (86-103) 99 (87-108) 64 (49-79)
FEV1 reversibility (%) 2.9 (1.8-8.0) 13 (11-19) 10 (2.4-18) 14 (4.2-26)
PEFR (% predicted) 105 (97-114) 98 (89-107) 96 (85-101) 70 (58-82)PEFR variability (%) N/A 17 (10-25) 21 (16-32) 19 (11-24)PD20 (mg methacholine) 0.19 (0.05-0.79) 0.25 (0.06-0.73)
Exhaled nitric oxide (ppb, at 50L/s) 15 (10-24) 52 (27-107) 29 (15-51) 19 (13-38)ClinicalAtopy (Skin prick positive, Y/N) 0 / 20 15 / 0 19 / 3 12 / 5
No. of skin prick allergens positive 0 (N/A) 6 (4-7) 3 (1.8-5) 4 (0-6)
Peripheral eosinophil count (109/L) 0.1 (0.1-0.2) 0.1 (0.1-0.6) 0.2 (0.1-0.3) 0.2 (0.1-0.2)Total IgE (iu/ml) 19 (10-57) 172 (21-451) 112 (45-189) 96 (24-526)
Body mass index (kg/m2) 25.6 (23.5-29.1) 23.6 (22.7-26.5) 25.1 (23.2-31.2) 33 (27.6-41.3)Smoking status
Never 17 14 18 12Former (Mean pack years) 3 (6.5) 1 (6.7) 4 (5.6) 4 (27)Current (Mean pack years) 0 0 0 1 (49)
Duration of asthma (years) N/A 18 (15-26) 22 (11-29) 41 (17-51)ACQ score N/A 0.6 (0.45-1.3) 1.0 (0.53-1.5) 2.7 (2.2-3.4)GINA level of control
Controlled N/A 8 (53) 4 (18) 0 (0)Partly controlled N/A 6 (40) 15 (68) 1 (5.9)Uncontrolled N/A 1 (6.7) 3 (14) 16 (94)
TreatmentInhaled steroids No No Yes Yes
Dose (equivalent mcg BDP) N/A N/A 400 (200-500) 1600 (1280-2040)Maintenance oral steroids (Y,N) No No No 4 / 13
Mean dose if taken (mg prednisolone/day) 8.3Long acting β agonist (Y/N) No No 9 / 13 17 / 0Leukotriene receptor antagonist (Y/N) No No 1 / 21 13 / 4Step on BTS treatment algorithm N/A 1 2 - 3 4 - 5
Inflammatory subtype (n,%)Neutrophilic 3 (21) 3 (23) 2 (12) 9 (5.3)Eosinophilic 1 (7.1) 2 (15) 3 (18) 4 (26)Mixed granulocytic 0 (0) 0 (0) 0 (0) 1 (5.9)Paucigranulocytic 10 (71) 8 (62) 12 (71) 3 (18)
Sputum cell differential (%)Macrophages 52 (31-69) 49 (35-64) 49 (30-63) 30 (18-41)Neutrophils 24 (8.0-64) 34 (22-54) 33 (15-51) 68 (39-78)Epithelial 4.9 (2.0-28) 4.3 (1.7-10) 5.3 (1.4-22) 1.1 (0.0-6.3)Eosinophils 0.38 (0.0-1.0) 1.5 (0.75-1.8) 0.75 (0.25-1.6) 0.69 (0.0-4.8Lymphocytes 0.1 (0.0-0.81) 0.25 (0-0.75) 0 (0.0-0.58) 0.0 (0.0-0.25)
BAL cell differential (%)Macrophages 79 (74-89) 70 (60-80) 81 (72-88) 72 (46-95)Neutrophils 2.8 (1.0-6.0) 2.5 (1.6-4.8) 3.5 (1.8-7.0) 5.5 (1.3-22)Epithelial 10.3 (4.0-19) 21 (13-35) 12 (7.0-19) 6.0 (2.8-11)Eosinophils 0.5 (0.0-0.75) 2.0 (0.75-3.6) 1.0 (0.6-3.3) 0.0 (0.0-1.0)Lymphocytes 1.5 (1.0-3.0) 1.5 (0.38-3.0) 1.5 (0.57-2.4) 1 (0.0-1.5)
Relevant comorbidities (n,%)Allergic rhinitis 0 (0) 12 (80) 10 (45) 7 (41)Nasal Polyps 0 (0) 0 (0) 1 (4.5) 4 (24)Eczema 2 (10) 7 (46) 6 (27) 3 (18)Bronchiectasis (history or CT) 0 (0) 0 (0) 1 (4.5) 1 (5.9)
Values are medians with interquartile ranges, unless stated otherwise. N/A: not available.Inflammatory subtype is based on sputum differentials using cut-points as per Simpson, J. L., R. Scott, et al. (2006). Respirology 11(1): 54-61 (neutrophilic: >61% neutrophils, eosinophilic: >3%). Percentages are of those with valid data.
ACQ, asthma control questionnaire; BDP, beclometasone dipropionate; BTS, British Thoracic Society; CT, computed tomogram; FEV1,
forced expiratory volume in 1 second; FVC, forced vital capacity; GINA, Global Initiative for Asthma; PEFR, peak expiratory flow rate; PD20, provocative dose 20.
Healthy controls Mild asthma Moderate asthma Severe asthma
Negative (>1.5) Not done
Timothy SC Hinks 5. MAIT cells – new players in asthma
158
Analysis of MAIT cells in human asthma
First, I studied the frequencies of MAIT cells in peripheral blood, induced sputum, BAL and bronchial
biopsies using flow cytometry. As described in the methods chapter, MAIT cells were defined as
CD3+ live lymphocytes (i.e. live T-cells) expressing the NK marker CD161 and the TCR Vα7.2 chain
(Figure 2.6). Whilst canonical MAIT cells are defined by their expression of the invariant Vα7.2-Jα33
TCR rearrangement, it has previously been shown that in humans surface expression of either CD161
or IL-18R, together with the Vα7.2 segment unequivocally identifies MAIT cells in both peripheral
blood and tissues (Martin, Treiner et al. 2009; Le Bourhis, Martin et al. 2010; Dusseaux, Martin et al.
2011). This definition enabled me to enumerate and to sort live MAIT cells based on their surface
staining properties alone.
MAIT cells are deficient in human asthma and correlate with disease severity
Vα7.2+CD161+ (MAIT) cells were abundant in airway tissues, comprising a median 1.8% (IQR 0.73-
3.0%) of T cells in health in blood, sputum, BAL and biopsy. There was no evidence of specific tissue
compartmentalisation (Kruskal-Wallis P=0.7). Frequencies of MAIT cells were lower in asthma than in
health in blood (P=0.005), in sputum (P=0.002) and in bronchial biopsies (P=0.02), with a similar
pattern in BAL (ns) (Figure 5.1). Furthermore when asthmatic subjects were stratified according to
three categories of disease severity - mild, moderate or severe asthma – there was a strong linear
trend across groups in PBMC (P<0.0001) and sputum (P=0.006, Figure 5.2), implying that this
deficiency correlated with disease severity.
Figure 5.1 MAIT cells are deficient in asthma
Frequencies of Vα7.2+CD161+ (MAIT) cells as a proportion of total live CD3+ T cells in peripheral
blood, sputum, BAL and biopsies in health and asthma. Box and whisker plots show medians and
IQRs. Differences are compared by unpaired t tests on Ln transformed data.
healthy controls n=21 PBMC, 13 sputum, 20 BAL, 14 biopsies.
asthma n=48 PBMC, 31 sputum, 40 BAL, 27 biopsies.
Timothy SC Hinks 5. MAIT cells – new players in asthma
159
Figure 5.2 MAIT cell deficiency correlates with asthma severity
Frequencies of Vα7.2+CD161+ (MAIT) cells as a proportion of total live CD3+ T cells in peripheral
blood, sputum, BAL and biopsies stratified according to disease severity. Box and whisker plots show
medians and IQRs. Linear trends are compared across groups using residuals on Ln transformed
data. (Not significant for BAL or biopsy).
healthy controls n=21 PBMC, 13 sputum, 20 BAL, 14 biopsies.
mild asthma n=14 PBMC, 11 sputum, 14 BAL, 10 biopsies.
moderate asthma n=17 PBMC, 10 sputum, 14 BAL, 10 biopsies.
severe asthma n=17 PBMC, 10 sputum, 12 BAL, 7 biopsies.
Timothy SC Hinks 5. MAIT cells – new players in asthma
160
Figure 5.3 Frequencies of a non-MAIT T cell subset do not differ in asthma
Frequencies of non-MAIT T cell subset Vα7.2+CD161- as a proportion of total live CD3+ T cells in
peripheral blood, sputum, BAL and biopsies stratified according to disease severity. Box and whisker
plots show medians and IQRs. No differences between groups were statistically significant.
healthy controls n=21 PBMC, 13 sputum, 20 BAL, 14 biopsies.
mild asthma n=14 PBMC, 11 sputum, 14 BAL, 10 biopsies.
moderate asthma n=17 PBMC, 10 sputum, 14 BAL, 10 biopsies.
severe asthma n=17 PBMC, 10 sputum, 12 BAL, 7 biopsies.
By contrast, the frequency of Vα7.2+ CD161- T-cells in blood or tissue showed no correlation with
either the presence or the severity of asthma (Figure 5.3). These “non-MAIT” cells represent
conventional adaptive T cells which use differently rearranged TCR Vα7.2 segments, but not the
invariant Vα7.2-Jα33 rearrangement that is unique to the MAIT TCR. Hence, these results show a
selective reduction in both peripheral blood and tissue MAIT in asthma and they suggest that the
reduction in MAIT cells correlates with asthma severity.
MAIT cell frequencies are not related to age
In the light of this association with disease severity I investigated whether MAIT cell frequencies might
correlate with other clinical factors. One important factor to consider was the effect of age as iNKT cell
frequency in peripheral blood decreases with age, especially after the age of 45-50 years, in humans.
Whilst I had selected healthy controls which were age-matched to the mild and moderate asthmatic
subgroups, the severe asthmatic cohort in my study tended to be older as a result of my specific
interest in severe neutrophilic asthma. To confirm or refute the hypothesis that MAIT cell frequencies
Timothy SC Hinks 5. MAIT cells – new players in asthma
161
fall with advancing age I recruited a further cohort of 12 older healthy controls with a median age of 53
years (IQR 48-57) and enumerated MAIT cells in peripheral blood. There was no difference in MAIT
cell frequencies when compared to the younger healthy controls (median age 27, IQR 24-34 years,
n=15)(P=0.4, Figure 5.4).
Figure 5.4 MAIT cell frequencies are not related to age
Peripheral blood frequencies of Vα7.2+CC161+ MAIT cells compared in younger healthy individuals
(median age=27 years (IQR 24-34), n=15), older healthy individuals (53 years (48-57), n=12) and all
asthmatics (38 years (25-51), n=40). MAIT frequencies did not differ between the younger and older
healthy controls, (unpaired t test, P=0.4).
Timothy SC Hinks 5. MAIT cells – new players in asthma
162
Timothy SC Hinks 5. MAIT cells – new players in asthma
163
Figure 5.5 Clinical correlates of peripheral blood MAIT cell frequencies
Peripheral blood frequencies of Vα7.2+ CD161+ (MAIT) cells are correlated negatively with (A)
treatment step on BTS treatment algorithm, (B) level of asthma control according to GINA
classification, (C) level of asthma control according to asthma control questionnaire score, (D)
duration of asthma; and are correlated positively with lung function including (E) FEV1 and (F) PEFR
as percentages of predicted values. Figures present Spearman’s correlations. MAIT cell frequencies
are expressed as a percentage of total live CD3+ T cells. Frequencies of Vα7.2+ CD161+ (MAIT) cells
are correlated positively with (G) exhaled nitric oxide levels and (H) with the presence of allergic
rhinitis. (H) P value is for a Mann-Whitney test.
Clinical correlations with MAIT cell frequencies
Next I investigated other potential clinical correlates of MAIT cell frequencies with univariate analyses
using Spearman’s correlations. Amongst asthmatic subjects MAIT cell frequencies correlated
negatively with the advancement of treatment steps on the BTS treatment algorithm, being lowest in
step 4 and 5 subjects (Figure 5.5 A) (rs= -0.585, P<0.0001). MAIT cell frequencies also correlated
negatively with level of asthma control according to the GINA classification (Figure 5.5 B), being
lowest in uncontrolled asthma (rs= -0.548, P<0.0001), or according to score on the asthma control
questionnaire (Figure 5.5 C) (rs= -0.472, P=0.0003).
MAIT frequencies also correlated negatively with duration of asthma, being lowest in those with the
most long-standing asthma (Figure 5.5 D) (rs= -0.460, P=0.0006). Patients with the longest duration of
asthma tended to be older. However, as explained above, age alone as a potential confounding factor
is unlikely to explain these differences.
Timothy SC Hinks 5. MAIT cells – new players in asthma
164
MAIT cell frequencies correlated positively with lung function including percentage of predicted FEV1
(rs=0.379, P=0.005, Figure 5.5 E), and percentage of predicted peak expiratory flow (rs=0.362,
P=0.008, Figure 5.5 F), although the strength of the correlation was lower than the above mentioned
correlations. These findings are consistent with the observed inverse correlation of MAIT cell
frequency with disease severity.
Furthermore, I found weak positive correlations between MAIT frequencies and exhaled nitric oxide
(eNO; rs=0.344, P=0.01, Figure 5.5 G), and also with the presence of allergic rhinitis (P=0.01, Figure
5.5 H). This suggested the possibility that MAIT deficiency might relate to specific endotypes, i.e.
eosinophilic or neutrophilic subtypes of asthma, but this hypothesis was not supported by either
sputum or BAL cell differentials. The weak positive correlation of MAIT frequencies with allergic
rhinitis is therefore more likely to be a consequence of the relatively higher rates of allergic rhinitis in
the milder asthmatic subjects (Table 5.1).
Modulation of MAIT cell frequencies by corticosteroids
Whilst the associations I have observed with disease severity are strong, they may be confounded by
an effect of steroid therapy. The possibility that steroids may negatively modulate frequencies of MAIT
cells is also raised by the association with BTS treatment step and possibly also by correlation with
eNO, as high eNO levels are associated with lower corticosteroid use (McNicholl, Stevenson et al.
2012). I therefore correlated MAIT cell frequencies with the doses of inhaled corticosteroids (ICS) in
all subjects. MAIT cell frequencies were indeed negatively correlated with the dose of ICS in both
PBMC (rs= -0.584, P<0.0001, Figure 5.6 A), and to a lesser extent in BAL (rs= -0.315, P=0.048, Figure
5.6B). However, a causal link cannot be proven by simple correlation, so to confirm directly whether
corticosteroids can modulate MAIT cell frequencies I conducted an additional sub-study in which I
analysed MAIT cell frequencies before and after introduction of either low-dose inhaled corticosteroids
or higher-dose systemic corticosteroids.
Timothy SC Hinks 5. MAIT cells – new players in asthma
165
Figure 5.6 MAIT cell frequencies and use of inhaled corticosteroids
Frequencies of Vα7.2+ CD161+ (MAIT) cells are correlated negatively with daily dose of inhaled
corticosteroids (ICS) in (A) peripheral blood and (B) BAL. Figures present Spearman’s correlations.
BDP, beclometasone dipropionate.
Timothy SC Hinks 5. MAIT cells – new players in asthma
166
Figure 5.7 MAIT cell and non-MAIT cell frequencies before and after inhaled corticosteroids
Frequencies of MAIT (Vα7.2+CD161+) cells and a non MAIT cell population (Vα7.2+CD161- cells) in
12 steroid-naïve subjects before and after 7 days of treatment with 200 mcg bd inhaled Qvar. No
differences are significant by paired t tests.
Inhaled corticosteroids
I administered 200 mcg of inhaled ultrafine particle hydroluoroalkane-134a (HFA) beclometasone
dipropionate (Qvar) twice daily for 7 days to 12 steroid-naïve subjects with mild asthma. No significant
differences were observed after ICS therapy in frequencies of MAIT cells or a non-MAIT cell
population (Vα7.2+CD161- cells) in either peripheral blood or in sputum (Figure 5.7).
Oral corticosteroids
I hypothesised that modulation of MAIT cell frequencies might occur only with higher, systemic doses
of corticosteroids and so I conducted a second sub-study in which I measured peripheral blood MAIT
cell frequencies before and after 7 days treatment with 20 mg prednisolone once daily, orally. At this
dose there was a significant 23% decrease in median MAIT cell frequencies over the week
(P=0.03)(Figure 5.8 A). This implies that steroids can modulate frequencies of MAIT cells.
Timothy SC Hinks 5. MAIT cells – new players in asthma
167
Furthermore this modulation is specific to MAIT cells, as there was no change in the frequencies of
the comparator non-MAIT cell population (Figure 5.8 B).
Figure 5.8 MAIT cell and non-MAIT cell frequencies before and after oral corticosteroids
Frequencies of MAIT (Vα7.2+CD161+) cells and a non MAIT cell population (Vα7.2+CD161- cells) in
12 moderate asthmatic subjects, usually controlled on inhaled corticosteroids, before and after 7 days
of treatment with 20 mg od oral prednisolone. Frequencies are a % of live CD3+ T cells. P values are
for paired t tests.
Although it cannot be excluded that this steroid effect might be the sole explanation for the deficiency
of MAIT cells in more severe asthma, it is unlikely: In the majority of subjects the main route of steroid
administration was by inhalation, and yet the correlation between ICS dose and MAIT cell frequencies
was much stronger in PBMC than in BAL (Figure 5.6) or biopsy or sputum (correlations are not shown
as they were not statistically significant). If steroids were the sole driver for MAIT cell suppression
then it would be expected that the effect would be most marked on airway cells. Furthermore the data
presented in Figure 5.5 show that the correlation of MAIT frequencies with measures of asthma
control (B, C) is just as strong as their correlation with treatment (Figure 5.5 A and Figure 5.6 A),
which would not be expected if steroids were the only modulating factor. In the following section I will
present evidence for at least one other factor which could modulate MAIT cell frequencies in vivo.
Timothy SC Hinks 5. MAIT cells – new players in asthma
168
Seasonal variation in MAIT cell frequencies
Rationale for investigating seasonal variation
I carefully inspected my data-set to search for evidence of other modulating factors. It has recently
been documented that vitamin D3 is important for development and function of iNKT cells, another
innate-like T cell. The main supply of active Vitamin D3 for the body is provided through sunlight-
induced synthesis of Vitamin D in the skin, while food-related uptake of Vitamin D plays a very minor
role. Deficiency of vitamin D in utero in mice results in a significant reduction in iNKT which persists
life-long and is refractory to later vitamin D supplementation because it results from increased
apoptosis of early iNKT cell precursors in the thymus (Yu and Cantorna 2011). Furthermore, this
modulation has been linked to experimental airways hyper-reactivity (AHR) in mice, as vitamin D
receptor (VDR) knock-out prevents the development of AHR in a manner which can be rescued by
adoptive transfer of VDR competent cells (Yu, Zhao et al. 2011). To date, there have been no
publications relating vitamin D and MAIT cells. I therefore wondered whether there was any evidence
in my dataset of an association between peripheral blood MAIT cell frequencies and either the month
of subject birth (if there were a long lasting effect) or the month in which study samples were taken (in
case there were a short term effect on vitamin D levels).
Seasonal variations in MAIT cell frequencies
There was a strong association between the season in which phlebotomy was performed and the
MAIT cell frequency (ANOVA on Ln transformed data, P<0.001, Figure 5.13 A). MAIT cell frequencies
peaked in August with a nadir in February (Figure 5.13 B) and post-hoc tests revealed both summer
and autumn frequencies differed significantly from the Jan-Mar quarter (Figure 5.9 A), which was
consistent with my hypothesis that the lowest MAIT frequencies would be observed in the winter when
vitamin D levels would are at their lowest. This effect was specific to MAIT cells as it was not
observed with a non-MAIT population (Vα7.2+CD161- cells, ANOVA P>0.05).
Timothy SC Hinks 5. MAIT cells – new players in asthma
169
Timothy SC Hinks 5. MAIT cells – new players in asthma
170
Figure 5.9 Annual variation in MAIT cell frequencies
Peripheral blood MAIT cell frequencies vary over the course of the year and are highest in the
summer months. Figure shows log transformed Vα7.2+CD161+ cell frequencies from healthy and
asthmatic subjects according to the quarter in which phlebotomy was performed. (A) A sinusoidal
regression line is fitted, with a value for R2 of 0.16. ANOVA P<0.0001 with post hoc Dunnett’s *
P<0.05 and **P<0.01 compared with Jan-Mar. (B) The same data as A but presented according to
month in which sample was taken. R2=0.14. (C) The same data as A stratified by disease severity.
Plots show medians and IQR. P values are for ANOVA on Ln transformed data.
Sinusoidal regression yielded a value for R2 of 0.16 which means that 16% of the variance in MAIT
cell frequencies is attributable to this seasonal variation, which is impressive given the likelihood that
many other factors are also likely to modulate frequencies of MAIT cells.
I investigated whether this effect could be a result of confounding by non-random sampling of my
population: perhaps I sampled more severe asthmatics in winter months and healthy individuals in
summer months. This turned out not to be the case because when the data were stratified by disease
severity there was no evidence of such a systematic error, and I observed the same pattern in
healthy, mild and moderate individuals. This effect was significant in mild (P=0.016) and moderate
(P=0.035) subjects despite the smaller subgroup sizes (Figure 5.9 C).
Timothy SC Hinks 5. MAIT cells – new players in asthma
171
Development and characterisation of MAIT cell clones
The data presented above suggest that MAIT cells may be very relevant to respiratory immunology,
being abundant in the airways, related to airways disease and modulated by asthma therapies. I
have, therefore, begun to further investigate the biology of MAIT cells by establishing cell lines
(clones) which in turn enabled me to begin to investigate the functional capabilities of these cells. I will
describe the cloning technique, the validation of the TCR sequence of these clones and preliminary
data regarding the cytokine expression profile of these clones.
Cloning technique
With help from a post-doctoral fellow, Dr Salah Mansour, working with Professor Gadola, I adapted a
protocol developed for cloning iNKT cells (Matulis, Sanderson et al. 2010) and established 7 cell lines
from peripheral blood MAIT cells. This is the first time MAIT clones have been established in this
manner, although Gold et al. have generated MAIT cell clones in the presence of dendritic cells
infected with live MTB and rhIL-2(Gold, Cerri et al. 2010). The details of the protocol I used are
described in the methods section. Cloning efficiency was low at 1/80. I investigated whether cloning
efficiency would be improved by the addition of the Src family tyrosine kinase inhibitor dasatinib,
which can prevent activation-induced TCR and co-receptor down-regulation without inducing
apoptosis (Weichsel, Dix et al. 2008). Cloning efficacy was unaffected by dasatinib: clones were
established in 3/240 wells in the presence of dasatinib and 3/240 in its absence.
The surface phenotypes of these MAIT cell clones are shown in Figure 5.9. All clones expressed the
Vα7.2 TCR and CD161, but varied in their expression of the CD4 and CD8 co-receptors. The original
description of MAIT cells suggested that they were predominantly CD4-8- or CD8αα cells(Tilloy, Di
Santo et al. 1999), but others have subsequently shown them to be more commonly CD8+ (Turtle,
Swanson et al. 2009; Gold, Cerri et al. 2010; Walker, Kang et al. 2012) and to also include CD4+
subsets. Of the clones I established four were CD4+, two were CD8+, and one was CD4-8-.
Timothy SC Hinks 5. MAIT cells – new players in asthma
172
Figure 5.10 Surface phenotype of MAIT clones
Surface expression of CD4, CD8, CD161 and TCR Vα7.2 on the first 7 successful MAIT cell clones.
All clones are Vα7.2+ and CD161+, although ex vivo stimulation causes significant CD161
downregulation in many cells. Clones differ in their CD4 and CD8 expression profiles.
Timothy SC Hinks 5. MAIT cells – new players in asthma
173
Figure 5.11 Confirmation that MAIT clones express the invariant Vα7.2-Jα33 TCR
rearrangement
Clones which were selected for expression of TCR Vα7.2 and CD161 were analysed by PCR and gel
electrophoresis. Growing clones were resuspended, washed twice with PBS RNA extracted using
chloroform, reverse transcribed using qScript kit and cDNA amplified by PCR using the Bioline Taq kit
and primers specific for the canonical Vα7.2-Jα33 TCR rearrangement. cDNA was run on a 1.1%
agarose gel at 80 Volts for 30 minutes. A clear product of appropriate length over 400bp is seen in the
lanes for clones 1,3,4,5,7 and 8 implying these are true MAIT cells expressing the invariant Vα7.2-
Jα33 rearrangement. From left to right: positive control Jα33 cDNA, clones 6, 8, 3, 4, 5, 7, 1, cDNA
from an iNKT clone as negative control, no template RT control; no template PCR control, ladder.
Timothy SC Hinks 5. MAIT cells – new players in asthma
174
Confirmation of MAIT clones by PCR
To confirm that these established TCRVα7.2+ cell lines expressed the full Vα7.2-Jα33 TCR
rearrangement, I used RT-PCR to measure expression of mRNA for the MAIT TCR using primers
specific for the full Vα7.2-Jα33 segment. Gel electrophoresis of the PCR product revealed distinct
bands of over 400 base pairs from 6 of the clones (Figure 5.11) confirming presence of the canonical
rearrangement. Positive controls included cDNA for the Jα33 segment and also PBMC from a subject
known to have very high frequencies of MAIT cells. Negative controls included no template controls
from the RT and PCR steps. An iNKT clone was also included as a negative control.
Clone phenotype
Finally I carried out an initial investigation into the functional capacity of these putative MAIT clones by
measuring staining for intracellular cytokines after 4 hours ex vivo stimulation with PMA and
ionomycin (see representative FACS plots in Figure 5.12). All clones were strong producers of TNFα
but differed in their secretion of other cytokines (Figure 5.13). Some clones were strong producers of
the TH17 cytokine IL-17, others of the TH1 cytokine IFNγ, whilst other clones produced neither
cytokine but instead produced the TH2 cytokine IL-13. This suggests that peripheral blood MAIT cells,
similar to peripheral blood iNKT cells, are a functionally heterogeneous population.
Figure 5.12 Typical intracellular cytokine expression by a stimulated MAIT clone
Representative example of intracellular cytokine staining for IL-17, IFNγ, TNFα, IL-4, -5, -10, -13 and -
22 on the MAIT clone 3 after 4 hours ex vivo stimulation with PMA and ionomycin.
Timothy SC Hinks 5. MAIT cells – new players in asthma
175
Figure 5.13 Heterogeneity of cytokine expression profile of MAIT clones
Differing cytokine expression profiles in 6 MAIT clones stimulated ex vivo for 4 hours with PMA and
ionomycin measured by intracellular staining for IL-17, IFNγ, TNFα, IL-13. Plots show proportions of
cells staining positive for each cytokine as a percentage of total live CD3+ cells.
Timothy SC Hinks 5. MAIT cells – new players in asthma
176
Discussion
The data presented in this chapter comprise the first description of human MAIT cells in a respiratory
disease. Prior research on MAIT cells has focussed on peripheral blood or the GI mucosa where they
were originally described and, to date, the only published data on MAIT cells in the human lung was
the report that CD8+Vα7.2+ cells were detectable in the lymph nodes and lung parenchyma of organs
from 2 individuals which had been rejected for organ transplantation (Gold, Cerri et al. 2010). Here, I
have characterised these T-cells systematically in 74 subjects, in peripheral blood and a range of
airway tissues, and this led to the discovery that MAIT cells are deficient in asthma. I have explored
clinical correlates of these immunological findings in a clinical cohort whose clinical phenotypes have
been characterised in great detail, and explored the association of MAIT frequencies with therapeutic
use of corticosteroids in two intervention studies. Finally, my studies have revealed a strong seasonal
variation of MAIT cell frequency suggesting a role for Vitamin D3 in MAIT cell development.
Although MAIT cells have never been studied in asthma before, an association between MAIT cells
and asthma would not be unexpected. MAIT cells share many similarities with another invariant T cell
subset, iNKT cells (Treiner and Lantz 2006), which have been implicated in allergic airways disease in
several murine models (Hachem, Lisbonne et al. 2005; Meyer, DeKruyff et al. 2008; Pichavant, Goya
et al. 2008; Wingender, Rogers et al. 2011; Yu, Zhao et al. 2011). However, it has not been possible
to extrapolate these findings to human asthma (Mutalithas, Croudace et al. 2007; Vijayanand,
Seumois et al. 2007; Thomas, Chyung et al. 2010), which may be related to the much lower
abundance of iNKT cells in humans compared with mice (Treiner, Duban et al. 2005). Conversely the
much higher abundance of MAIT cells in humans (Treiner, Duban et al. 2005) suggests that MAIT
cells may perhaps fulfil a corresponding role to the murine iNKT.
Whilst I have demonstrated that steroids can modulate MAIT cell frequencies, as I have argued
above, it is unlikely that use of ICS is the sole mechanism for MAIT deficiency in asthma. The
correlation between ICS dose and MAIT frequency is strongest in peripheral blood rather than the
airway tissues where drug delivery occurs. MAIT cell frequencies correlated just as strongly with
measures of asthma control as with steroids doses. Furthermore, it remains to be determined whether
the steroid-induced suppression of MAIT cells is beneficial or detrimental to the integrity of the
mucosal immune system. In a murine model of inflammatory colitis MAIT cells have been
demonstrated to play a protective, anti-inflammatory role in the GI mucosa (Ruijing, Mengjun et al.
2012). Moreover, current understanding of MAIT cells suggests that they are an important mechanism
in preventing bacterial and mycobacterial infections (Gold, Cerri et al. 2010; Le Bourhis, Martin et al.
2010; Kjer-Nielsen, Patel et al. 2012). Therefore, steroid-induced suppression of MAIT cells might
underlie the increased risk of invasive pneumococcal disease associated with severe asthma (Talbot,
Hartert et al. 2005; Klemets, Lyytikainen et al. 2010) and increased risk of pneumonia in subjects with
COPD receiving inhaled fluticasone (Calverley, Anderson et al. 2007; Crim, Calverley et al. 2009;
Welsh, Cates et al. 2010). Given the very widespread use of ICS, this possibility certainly warrants
Timothy SC Hinks 5. MAIT cells – new players in asthma
177
further investigation. Such investigations should also explore potential deficiencies in MAIT cell
function.
My finding of a link between seasonality and MAIT cell frequencies also warrants further investigation.
This was a novel, hypothesis-driven observation and is consistent with what is already known about
the existence of mechanisms by which invariant T cell number and function can be modulated by
vitamin D (Yu and Cantorna 2011; Yu, Zhao et al. 2011). If this link were confirmed then it would add
to our understanding of the relationships between asthma and vitamin D, which is currently an area of
controversy. Studies in humans have provided evidence that vitamin D can modulate T cell immunity
in vivo. Vitamin D levels are positively correlated with peripheral blood TH1/TH2 ratios and to a lesser
extent with Treg frequencies in asthma (Chambers, Nanzer et al. 2012; Maalmi, Berraies et al. 2012).
In vitro free 25(OH)2 vitamin D3 influences the balance between inflammatory and regulatory T cell
responses by its effect on dendritic cells (Jeffery, Wood et al. 2012). However large scale clinical
studies have not yet provided a clear understanding of the significance of these findings. Children with
asthma seem to be at increased risk of vitamin D deficiency and there are associations between low
vitamin D and worse asthma symptoms, disease severity, frequencies of exacerbations and poorer
lung function (Gupta, Bush et al. 2012). However, most data come from case control studies which
have provided conflicting results and are affected by selection bias (Paul, Brehm et al. 2012).
Furthermore, the vitamin D effect may be weaker in adults (Goleva, Searing et al. 2012). Several
longitudinal studies are now ongoing, the first of which found that low vitamin D levels were
associated with increased risk of asthma exacerbations in over 1000 children with mild-to-moderate
asthma (Wu, Tantisira et al. 2012).
The seasonal variation I have observed could be secondary to factors other than sunlight exposure-
related vitamin D levels. In particular, it could be that the winter nadir in MAIT cells is due to more
frequent viral exacerbations or increased use of oral steroids at that time of year. Arguing against this
would be the observation that the seasonal variation was most marked in the mildest subjects, who
were the least likely group to suffer from viral exacerbations or require oral steroids. To resolve this
uncertainty I have identified 87 serum samples which are temporally paired with the peripheral blood
MAIT cell data and in which vitamin D metabolites are currently being measured by mass
spectrometry by my collaborators Prof Alan Jackson and Dr Steve Wootton in the Southampton
Nutrition BRC (see acknowledgements list).
The ability to establish and maintain MAIT cell lines is a valuable tool for the next steps in
investigating MAIT cell biology. My work provides proof of concept that clones can be established
without the need for infection of dendritic cells with live MTB, the only published method to date (Gold,
Cerri et al. 2010). The low cloning efficiency will need to be addressed in future work, which might
initially compare the efficiency of stimuli other than PHA, such as anti-CD3 OKT, microbial ligands or
heat killed BCG. However a possible explanation for this low cloning efficiency can be inferred from a
very recent publication which found that MAIT cells have a strong predisposition to apoptosis due to
Timothy SC Hinks 5. MAIT cells – new players in asthma
178
high expression of caspases 3 and 7 (Gerart, Siberil et al. 2012). These authors made this
observation by studying X-linked lymphoproliferative syndrome, a primary immunodeficiency caused
by mutation in the X-linked inhibitor of apoptosis (XIAP) which is a physiological inhibitor of caspases
3, 7 and 9 (Eckelman, Salvesen et al. 2006). This genetic disorder leads to a deficiency in frequencies
of iNKT and MAIT cells leading to a susceptibility to Epstien-Barr virus infection, haemophagocytic
lymphohistiocytosis and hypogammaglobulinaemia (Pachlopnik Schmid, Canioni et al. 2011; Gerart,
Siberil et al. 2012). Gerart et al. showed that this pro-apoptotic tendency can be reversed by inhibition
of the transcription factor PLZF/ZBTB-16 (Gerart, Siberil et al. 2012); thus potential avenues for
increasing this cloning efficiency might include inhibition of PLZF/ZBTB-16 or caspases 3 and 7. On
the other hand, addition of the Src family tyrosine kinase inhibitor dasatinib, which inhibits TCR-
mediated cell activation, did not improve cloning efficacy. However, it is possible, that – in the
absence of selective MAIT antigens – using the nonspecific mitogen PHA, which potentially crosslinks
many different glycosylated surface receptors, overrode the inhibitory effect of dasatinib on T-cell
activation.
Conclusions
The high abundance of MAIT cells - which I have found comprise an average of 2% of T cells in
peripheral blood and 4% of bronchial biopsy T cells in health – and their remarkable homology across
diverse mammalian species indicates that they serve an important function within the immune system.
As yet this function and the factors triggering a specific MAIT response (Kjer-Nielsen, Patel et al.
2012), remain poorly understood. My results are consistent with an important role of MAIT cells in the
airways, and together with published studies it is likely that they are key players during respiratory
host defence, including resistance to pneumonia, invasive bacterial disease, bronchiectasis and
opportunistic infections such as opportunistic mycobacterial infections. My data indicate that MAIT cell
homeostasis is markedly disturbed in asthma, particularly more severe forms and that this
disturbance is exacerbated by therapeutic use of corticosteroids. It remains to be seen whether this
has beneficial or detrimental consequences for the integrity of the respiratory mucosal immune
system, and whether or not seasonal variations in MAIT frequencies are directly related to Vitamin D
metabolism.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
179
CHAPTER 6
Deep sequencing of the airway microbiome Deus ex operibus cognoscitur 6
6 Sir Isaac Newton PRS MP (1642-1727) Translated ‘God is known from his works’. Isaac
Newton, Cambridge University Library MS Add. 3695, section 13
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
180
Introduction
Having systematically characterised both adaptive and innate-like T cell populations in asthma, I
undertook to investigate the presence of potential airway microbial stimuli which might be driving
these immune responses. The airways were once thought to be sterile in health (Laurenzi, Potter et
al. 1961; Pecora 1963), but there is accumulating evidence for the presence of bacteria in the airways
(Charlson, Bittinger et al. 2011) and perhaps even a commensal airway microbial community (Hilty,
Burke et al. 2010; Huang, Nelson et al. 2011). However, only two studies have been conducted to
date, and so the relevance of finding microbes in disease and health remains unknown. Some studies
suggest that bacteria or fungi may drive the recruitment of neutrophils to the airways in neutrophilic
airways disease (Simpson, Grissell et al. 2007; Green, Kehagia et al. 2008; Simpson, Powell et al.
2008; Essilfie, Simpson et al. 2012).
It is also important to consider the role of respiratory viruses in asthma. A strong causal link has now
been established between impaired innate response to acute viral infections and the development of
acute asthma exacerbations (Wark, Johnston et al. 2005; Contoli, Message et al. 2006; Message,
Laza-Stanca et al. 2008). Others have postulated that viruses may play a further aetiological role in
the pathogenesis of asthma, either through early life infections (Jartti, Paul-Anttila et al. 2009), or
possibly through impaired viral clearance leading to chronic viral persistence (Kling, Donninger et al.
2005; Harju, Leinonen et al. 2006; Wos, Sanak et al. 2008).
As introduced in chapter 1, the microbiome constitutes the totality of microbes, their genomes, and
environmental interactions in a particular environment (Highlander 2012). The emerging use of high
throughput molecular techniques to identify microbes without the need for traditional culture
techniques has transformed our ability to characterise the microbial flora in distinct anatomical niches
of the human microbiome. Until recently analysis of the respiratory tract flora has depended on
traditional culture-based microbiological techniques, which tended to suggest the airways were sterile
(Laurenzi, Potter et al. 1961; Pecora 1963). However, only 70% of body surface microbes (Han,
Huang et al. 2012) and only 1% of all known microorganisms can be cultured by such techniques
(Staley and Konopka 1985). Furthermore culture-based techniques are biased towards selecting for
organisms which grow on the chosen culture media, they typically excluded organisms normally
present at high levels in the upper respiratory tract, or restricted analysis to potential pulmonary
pathogens, and depended on arbitrary quantitative thresholds for clinically significant numbers of
organisms (Charlson, Bittinger et al. 2011).
More recently culture-independent techniques have been developed for microbial analysis with
several advantages. Rather than detecting a handful of species, they can instead characterise entire
microbial populations, involving much less bias and providing accurate measurements of relative
abundances of species(Charlson, Bittinger et al. 2011). To date studies of the human airway
microbiome have used PCR based methods to analyse bacterial 16s rRNA. In cystic fibrosis (Harris,
De Groote et al. 2007; Guss, Roeselers et al. 2011; Daniels, Rogers et al. 2012) and COPD (Han,
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
181
Huang et al. 2012; Sze, Dimitriu et al. 2012) these studies have shown that a much broader range of
species is present in the airways, many of which cannot be cultured or are anaerobic species not
previously thought to survive in the airways. Furthermore, they have identified associations between
distinct bacterial community compositions and particular disease phenotypes (Sze, Dimitriu et al.
2012). However, to date, only two published studies have used this technique to analyse the airway
microbiome in health(Hilty, Burke et al. 2010; Charlson, Bittinger et al. 2011), of which only one
compared the airway microbiome in health and asthma(Hilty, Burke et al. 2010). The careful analysis
of the airway microbiome in health by Charlson et al. has suggested a continuity of the lower airway
microbiome with bacterial communities found in the upper respiratory tract; this was in contrast to the
conclusion drawn by Hilty et al. of the existence of a core pulmonary microbiome comprising distinct
microbial communities. Hilty et al. also reported an increase in the presence of Haemophilus species
in asthma using bronchial brushings. This single study in asthma was small, involving only eight
healthy adults and 11 asthmatics, and may not be representative of most cases of asthma. In
addition, it did not account for recent antibiotic usage. Furthermore, 16s rRNA sequencing only
detects prokaryotic organisms (bacteria and archaea), but not viruses or fungi. Similarly, to date no
one has attempted to correlate the analysis of the airway microbiome in asthma with immunological
read-outs.
In contrast to these studies, and in order to measure the broadest possible range of species and to
minimise bias towards specific taxa, I elected to analyse microbial RNA and DNA in respiratory
samples by whole genome shot-gun sequencing (‘deep sequencing’) using the Roche/454 next-
generation sequencing platform. I chose this technique because, unlike the technologies mentioned
above, this can detect not only bacterial, but also fungal and viral genomes including those
incorporated into the human genome. The technique has very high sensitivity, and is ideal for
detecting previously unknown species as it does not require prior knowledge of the organisms
expected. Currently there are no publications in the literature using this technique on respiratory
samples.
Data were analysed using the VirusHunter analysis pipeline(Zhao) in which microbial sequences were
identified on the basis of Basic Local Alignment Search Tool (BLAST) alignments and the taxonomic
classification of the reference sequences to which a read is aligned. A ‘read’ is a short sequence
generated by high-throughput sequencing and typically <1000 base pairs in length. In phylogenetic
analysis a unique organism is referred to as an operational taxonomic unit (OTU). An OTU is defined
by the National Centre for Biotechnology Information (NCBI) as a ‘taxonomic level of sampling
selected by the user to be used in a study, such as individuals, populations, species, genera, or
bacterial strains’(Blaxter, Mann et al. 2005).
Pyrosequencing is a technique of massively parallel DNA sequencing capable of sequencing roughly
400-600 megabases of DNA per 10-hour run. RNA is extracted from cells and reverse transcribed into
cDNA. Genomic and cDNA are ligated to adapters and fixed to small DNA-capture beads in a water-
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
182
in-oil emulsion. The DNA fixed to these beads is then amplified by PCR. Each DNA-ligated bead acts
as a separate microreactor in which parallel DNA amplifications are performed, yielding approximately
107 copies of a template per bead (Margulies, Egholm et al. 2005). Each bead is then placed by
centrifugation into a 29 µm well on a fibre optic chip with smaller beads carrying a mix of enzymes
including DNA polymerase, ATP sulfurylase, and luciferase(Voelkerding, Dames et al. 2009).
Sequencing is based on the detection of pyrophosphate released during DNA synthesis, using a
cascade of enzymatic reactions in which visible light is generated in proportion to the number of
nucleotides incorporated and detected using a charge-coupled device(Ronaghi 2001; Margulies,
Egholm et al. 2005). Pyrosequencing differs from Sanger sequencing in that it detects incorporation of
pyrophosphate rather than chain termination with dideoxynucleotides, and has the advantages of
greater accuracy, parallel processing and automation (Ronaghi 2001).
Samples were collected by me but were sequenced and analysed by the Virgin Laboratory,
Washington School of Medicine at St Louis (see acknowledgment list).
Results and comments
Participants
Forty-seven BAL and 39 sputum samples were obtained from 55 individuals during periods of clinical
stability and at least 6 weeks after the end of the last known respiratory infection. These individuals
were a subset of the population reported in chapter 3 and comprised 9 with mild asthma, 16 with
moderate asthma, 15 with severe asthma and 15 healthy controls. Samples were sequenced and
analysed in two separate batches, generating a pilot data-set of 9 BAL samples and 8 sputum
samples, and a subsequent main data-set of 38 BAL samples and 39 sputum samples. The results
section is, therefore divided so as to present the bacterial and viral analyses separately. The first part
of the bacterial analyses comprises results from a) pilot study and b) main study. Further parts c) and
d) report on the pilot and main analyses conducted on sputum samples. The second section reports
on viral analyses and is also structured as pilot and main study.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
183
Results section I
a) Bacterial species in BAL from the pilot study
Potentially significant bacterial sequences from the pilot data-set are presented in Table 6.1. Bacterial
OTU identified by only a single read, or those with poor homology to the reference sequence, have
been excluded.
Table 6.1 Bacterial OTU identified from BAL in the pilot dataset.
Subject ID Classification OTU (species) Number of reads
Sequence homology (range, %)
104 Healthy control None
321 Moderate asthma Escherichia 2 98.6 - 99.5
403 Severe asthma Streptococcus mitis 2 61.8 – 92.1
Prevotella melaninogenica 2 84.6 – 97.8
404 Severe asthma None
406 Severe asthma None
407 Severe asthma None
409 Severe asthma None
412 Severe asthma Haemophilus influenzae 38 90.6 – 100
Leptotrichia buccalis 7 76.5 – 93.1
Environmental Eubacterium 6 97.9 - 100
422 Severe asthma Tropheryma whipplei 35 80.3 – 100
Rothia mucilaginosa 3 97.6 – 99.5
As seen in Table 6.1, in 4/9 individuals, including the healthy control, there was no evidence of
bacterial colonisation. In the other five individuals, there was evidence of the presence of bacteria
typical of those obtained from the oral cavity or upper respiratory tract: the gram-positive cocci Rothia
mucilaginosa and Streptococcus mitis and the anaerobic gram-negative bacilli Prevotella
melaninogenica and Leptotrichia buccalis. These would be consistent with microaspiration from the
upper respiratory tract as suggested recently (Gleeson, Eggli et al. 1997; Charlson, Bittinger et al.
2011). As with the study by Charlson et al. (Charlson, Bittinger et al. 2011) it is not clear whether
these sequences are derived from live or dead bacteria.
Two organisms were identified by ≥35 separate reads, implying much higher abundance, and are of
more specific interest. The gram-negative bacillus Haemophilus influenzae, which is an opportunistic
respiratory pathogen, was identified in one subject (412), a 63 year old male with severe neutrophilic
asthma and frequent exacerbations. In this particular instance, because of high clinical suspicion, the
lavage fluid was also sent for routine microbiological culture, which yielded heavy growth of H
influenza sensitive to amoxicillin, doxycycline and erythromycin. High resolution computed
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
184
tomography (HRCT) revealed mild bronchial wall thickening of the lower lobes with insufficient
evidence of bronchiectasis, whilst the sample originated from the right upper lobe which was
unaffected. There was marked neutrophilia in both induced sputum (71%) and BAL (68%). As a
consequence of the culture result, the subject was initiated on long term amoxicillin, to which he
responded well. Specifically in the 12 months prior to the bronchoscopy he had experienced 20
exacerbations requiring oral steroids. He was using reliever inhaler 3-4 per day and reported waking
3-4 times per night. After initiation of long term antibiotics he experienced a 1.1 point fall in ACQ,
suffered only one exacerbation in 18 months and reported in a typical day requiring no reliever
medication and having no nocturnal awakenings.
Deep sequencing has not previously been used as a clinical tool in respiratory medicine, but this case
study provides an interesting proof of principle that data obtained by culture-independent techniques
can correlate with both traditional culture results and with the clinical picture.
In the same individual the TH17 cell frequency in BAL was strikingly elevated at 11.3%. This was the
highest BAL TH17 cell frequency recorded in the whole study (n=60) and was well above the normal
range I have observed: median 2.59% (IQR 1.28-4.03%). By contrast peripheral TH17 frequencies
were not elevated, but at 0.47% were in the bottom quartile: median 0.58, (IQR 0.38-0.77%). This is
consistent with the hypothesised role of TH17 migrating out of peripheral blood and into, and
differentiating within, sites of early or ongoing microbial infection (Veldhoen and Stockinger 2006;
Torchinsky, Garaude et al. 2009). Thus, although this is a single case, I was able for the first time to
correlate immunological and metagenomic data.
The second organism identified at high abundance in this dataset was Tropheryma whipplei from
subject 422. This was detected in 35 separate reads with 99% nucleotide and 100% amino acid
homology to the reference database over their full length. BAL cytospins contained foamy
macrophages which are typical in Whipple’s disease but are non-specific. PAS staining of cytospins
and bronchial biopsies and standard culture of BAL were negative, so a definitive diagnosis of
pulmonary Whipple’s disease has not yet been made.
T whipplei is a gram positive actinobacteria which can cause a serious but rare systemic bacterial
infection affecting virtually any organ, with a wide variety of clinical presentations. Isolated lung
disease is rare(Urbanski, Rivereau et al. 2012), although recognised manifestations include chronic
cough, pleural effusions, hilar lymphadenopathy and pulmonary infiltrates including nodular shadows
and basal parenchymal interstitial infiltrates(Ratnaike 2000). Whipple’s disease is associated with
immune dysfunction including defects in intracellular killing by monocytes and macrophages, defects
in the interleukin-12 axis and alterations in lymphocyte populations(Ratnaike 2000; Schinnerling,
Moos et al. 2011) such as an increase in the TH2/T reg ratio(Biagi, Badulli et al. 2012).
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
185
Interestingly subject 422 has evidence of a longstanding abnormality of cell mediated immunity. She
is a 23 year old female who presented first to tertiary services at age 11 with frequent infective
exacerbations of severe eczema. She has since suffered from recurrent skin infections including
breast abscesses and chronic genital yeast infections, and has a long-term extremely elevated IgE,
with levels of over 46000 IU in the past. She is atopic with a history of peanut allergy. Her asthma
developed in early infancy and is again characterised by neutrophilia (sputum neutrophils 74%),
without bronchiectasis, and her predominant symptom is chronic cough. Although this clinical
presentation does not meet diagnostic criteria, (Woellner, Gertz et al. 2010) it is reminiscent of hyper
IgE syndrome (Job’s syndrome), which is associated with mutations in signal transducer and activator
of transcription (STAT)3 leading to insufficient expression of RORt and consequent deficiency of
TH17 cells (Holland, DeLeo et al. 2007; Ma, Chew et al. 2008; Milner, Brenchley et al. 2008). Subject
422 had normal TH17 cell frequencies in blood (0.87% compared with a study median of 0.52%, IQR
0.31-0.90%) but she had low TH17 frequencies in bronchial biopsies (2.3% compared with a study
median 3.3%, IQR 2.3-6.8%) and very low BAL TH17 frequencies at 0.1% (study median 2.6%, IQR
0.85%-4.0%) which is the second lowest value I have recorded. This raises the possibility that the
subject was predisposed to acquisition of T whipplei as a consequence of a primary pulmonary TH17
cell deficiency, for instance due to a defect in TH17 specific chemokines. Thus I have again provided
anecdotal evidence of a correlation between immunological and metagenomic data.
b) Bacterial species in BAL from the main study
Next sequencing of BAL was conducted on samples from a further 38 subjects. Unfortunately, my
collaborators unexpectedly added an additional processing step of passing the defrosted samples
through a 24 µm filter with the aim of increasing the relative abundance of viral sequences. This may
have decreased the relative number of bacterial and fungal reads making it impossible directly to
combine the pilot and definitive data-sets. Nonetheless, many bacterial reads were detected - typically
500 sequences but in some cases over 5000.
The proportions of bacterial taxa in each BAL sample are presented in Figure 6.1, in which subjects
have been arranged by hierarchical cluster analysis to emphasize taxa with similar abundance
patterns. It is apparent from this figure that no specific pattern emerges within the hierarchical
clustering of the subjects implying no evidence of association between disease and bacteria in this
data-set. One sample from a moderate asthmatic contained sequences from Acinetobacter and
Moraxella which, again, might represent microaspiration as these are recognised upper airway flora.
Hilty et al. reported an increase in the presence of proteobacteria particularly haemophilus species in
asthma using bronchial brushings (Hilty, Burke et al. 2010). I did not replicate this observation.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
186
Figure 6.1 Proportions of bacterial taxa in each bronchoalveolar lavage sample
Proportions of bacterial taxa in each sample inferred from pyrosequence data. Each column
corresponds to a specific bacterial order (A), family (B), or genus (C). Each row corresponds to an
individual subject. Columns and rows have been subjected to hierarchical cluster analysis to
emphasize taxa with similar abundance patterns. The proportional representation (relative
abundance) of each family is represented by the colour code. Subjects are identified by a 3 digit
number. The first number corresponds to the major disease classification: 1xx, healthy control; 2xx,
mild asthma; 3xx, moderate asthma; 4xx, severe asthma. It can be seen, therefore, that no specific
pattern emerges within the hierarchical clustering of the subjects, implying no evidence of association
between disease and bacteria in this data-set. Abundance was low for most taxa, with the following
exceptions (see text): Erysipelotrichales Erysipelotrichaceae Coprobacillus and Archoleplasmatales
Archoleplasmataceae Phytoplasma. In addition one sample (subject 313, a moderate asthmatic)
contained reads for Acinetobacter and Moraxella).
Otherwise the main finding was that abundance was low for most taxa, with the exceptions of two:
Erysipelotrichales Erysipelotrichaceae Coprobacillus and Archoleplasmatales Archoleplasmataceae
Phytoplasma. These taxa are not recognised respiratory flora. Coprobacillus are gram negative bacilli
found in human faeces(Lyra, Rinttila et al. 2009; Park, Kim et al. 2011), whilst Phytoplasma are plant
pathogens (Strauss 2009; Gasparich 2010) whose DNA might well be detectable in human faeces. In
the light of this and given the detection in nearly all BAL samples it is possible that these are
contaminants which have been transferred to the bronchoscopes during the cleaning process. Whilst I
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
187
took samples through a sterile BAL catheter (Combicath) to minimise contact between bronchoscopes
and lavage fluid, BAL fluid can still contact the tip of the scope where this forms the ‘wedge’.
Subsequent investigation revealed that the bronchoscopes were cleaned in a central facility which
also handle lower gastrointestinal endoscopes and which would be likely to carry a very high biomass
of faecal bacteria. It is possible that microbial DNA could have survived cleaning with the acidic
oxidising agent peracetic acid and been transferred to the bronchoscopes. Similar contamination with
a low background of soil- and water-associated organisms was noted by Charlson et al. in their pre-
bronchoscopy channel specimens (Charlson, Bittinger et al. 2011).
Figure 6.2 Bacterial abundance in bronchoalveolar lavage
Plot of raw number of bacterial sequence reads obtained from each bronchoalveolar lavage sample,
grouped according to disease classification (x) axis. Number of sequences obtained ranges from 5 to
over 5000, and is related to relative bacterial abundance. No significant differences were observed
between the distributions according to disease severity, even if outliers were removed. Note that
these plots do not contain data from an initial pilot data-set, which is why severe subjects are under-
represented.
Figure 6.2 shows the relationship between the raw number of bacterial sequences (which can be
considered a marker of relative bacterial abundance (Charlson, Bittinger et al. 2011)) obtained from
each bronchoalveolar lavage sample, according to disease classification. There were no statistically
significant differences between bacterial abundance and either presence or absence of asthma, or
disease severity within the asthma phenotype.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
188
c) Bacterial species in sputum from the pilot study
Deep sequencing was also conducted on a pilot set of nine sputum samples from subjects with
moderate asthma taken during acute upper respiratory tract infections. This revealed multiple
bacterial genomes in each of the exacerbation sputum samples as shown in Table 6.2. These are all
recognised dental or upper respiratory tract flora from the same microbial communities as those
observed at lower frequencies in the lung, notably Rothia mucilaginosa, Veillonella parvula,
Actinomyces, and Enterococcus, which are all typical oral flora(Vaccher, Cordiali et al. 2007; Bizhang,
Ellerbrock et al. 2011; Brittan, Buckeridge et al. 2012) or and bacteria such as Streptococcus
pneumoniae and Neisseria meningitides which colonise the upper airways(Brittan, Buckeridge et al.
2012). This similarity of bacteria between upper and lower airways would be expected from mixing
with saliva during sputum induction, or potentially from colonisation of central airway mucous by
bacteria micro-aspirated into the airways. In this study it is not possible to distinguish bacteria arising
from the airways from those which were present in saliva. Each of these bacterial families were also
observed by Charlson et al., namely Veillonella, Enterococcus, Neisseria spp,, Actinomycetales
micrococcaceae which includes Rothia mucilaginosa, Actinobacteria Actinomycetaceae which
includes the Actinomyces, and Firmicutes Lacobacillales Streptococcaceae which includes
Streptococcus pneumoniae (Charlson, Bittinger et al. 2011).
Matched samples from a single individual at two visits 3 days apart have been analysed from subject
53. Three out of 4 species detected on symptom day 7 were detected again on symptom day 10,
reflecting the consistency of the technique in detecting what are presumably fairly stable polymicrobial
populations of commensal flora.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
189
Table 6.2 Bacterial OTU identified by more than one read from sputum samples collected
during acute viral upper respiratory tract infections
Subject ID
Day of sample collection (symptom day 1 to 10)
OTU (species) Number of reads
Sequence homology (range, %)
10 7 Rothia mucilaginosa 321 80.0 - 100
Streptococcus sp 213 65.3 - 100
Actinomyces sp 169 74.2-100
11 4 Veillonella parvula 5 70.8 - 96.3
Neisseria meningitides 5 86.8 - 99.4
Streptococcus pneumoniae 14 82.6 - 98.1
Streptococcus mitis 15 61.9 - 97.4
27 1 Prevotella melaninogenica 119 67.5 - 100
Rothia mucilaginosa 45 72.0 - 100
Streptococcus sp 31 74.1 - 100
Veillonella parvula 28 69.8 - 99.3
Streptococcus mitis 25 70.0 - 99.1
30 7 Rothia mucilaginosa 20 91.4 - 100
Streptococcus sp 4 93.8 - 100
Streptococcus oralis 2 93.2 - 99.3
35 10 Rothia mucilaginosa 359 79.5 - 100
41 1 Prevotella melaninogenica 156 39.3 - 99.7
Veillonella parvula 55 63.6 - 99.7
Rothia mucilaginosa 29 45.0 - 100
Streptococcus sp 28 73.4 - 100
Rothia dentocariosa; 24 45.6 - 100
43 1 Haemophilus parainfluenzae 36 85.8 - 99.4
Rothia mucilaginosa 32 83.7 - 100
Streptococcus sp 26 68.6 - 100
Streptococcus pneumoniae 19 78.0 - 98.9
53 4 Streptococcus sp 101 67.2 - 100
Rothia mucilaginosa 96 82.9 - 100
Veillonella parvula 89 71.5 - 99.6
Actinomyces sp 31 79.8 - 99.1
53 7 Rothia mucilaginosa 211 71.6 - 100
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
190
Streptococcus sp 98 85.4 - 100
Enterococcus sp 33 76.4 - 96.4
Veillonella parvula 22 84.3 - 100
d) Bacterial species in sputum from the main study
Thirty-nine sputum samples were obtained from the same cohort of patients. These samples were
filtered through a 24 µm filter after thawing. Bacterial genomes were detected with higher abundance
in sputum samples, but indicated the presence of the same microbial communities as those observed
at lower frequencies in the BAL, notably Rothia mucilaginosa, Veillonella parvula, Actinomyces,
Enterococcus, which are all typical oral flora(Vaccher, Cordiali et al. 2007; Bizhang, Ellerbrock et al.
2011; Brittan, Buckeridge et al. 2012) and similar to those species observed by Hilty (Hilty, Burke et
al. 2010). Given the similarity to the BAL data a detailed analysis of individually annotated taxa was
not performed.
Summary
In summary, these analyses of the bacterial flora in BAL and sputum show the presence in the lower
airways of typical microbes of the oral and upper-respiratory tract, but have not shown evidence of
distinct lower airway microbial communities. Cladistic analysis does not suggest general differences in
microbial communities between asthma and health, with the exception of some individuals who may
have colonisation with a specific respiratory pathogen, whose presence may correlate with clinical
and immunological features.
Results section II
Viral species in sputum and BAL
Figure 6.3 shows a representation of the relative abundance of viral sequences in BAL according to
virus family and disease phenotype. All samples in this data-set were passed through a 24 µm filter
with the aim of enriching the abundance of viral sequences relative to those of human or bacterial
genomes. Retroviral sequences were almost universally present, which is expected due to the high
frequencies of endogenous retroviruses incorporated into the human genome. These comprise up to
8% of the human genome and do not indicate actively replicating viruses (Belshaw, Pereira et al.
2004; Bizhang, Ellerbrock et al. 2011). Similarly, the presence of low numbers of sequences identified
as poxvirus, phycodnavirus and iridovirus families is unlikely to be significant because these
sequences probably belong not to viruses but to the orthologous host sequence. This sequence
similarity can occur due to viral hijacking of host genes or due to the presence of shared repetitive
motifs (Handley 2012). As can be seen from this figure I observed no association between the
presence of viral sequences and the presence of asthma or the severity of asthma.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
191
Figure 6.3 Viral taxa in bronchoalveolar lavage samples
A plot of numbers of viral sequence reads for each individual in bronchoalveolar lavage, arranged
according to viral family. The diameter of each circle is proportional to the number of reads
sequenced in each individual. Individuals are arranged along the x axis, grouped according to disease
classification.
Despite using a sample preparation and data analysis pipeline which has been developed and
validated for the detection of novel viruses (Zhao), the analysis failed to detect evidence of any
particular virus at significant copy number in either BAL or sputum, from these cross sectional
samples.
One virus, Betatorquevirus, was repeatedly detected at low copy number in BAL from 2 subjects with
severe neutrophilic asthma receiving oral (subject 409) or high dose inhaled (subject 422)
corticosteroids, as well as 3/9 sputum samples from subjects with suspected acute viral exacerbations
of asthma. The significance of this finding is not clear. These samples were taken over a 5 month
period from one geographical area and it is possible one virus might have been circulating in the
community. Little is known about Betatorquevirus, also known as Torque Teno Mini Virus (TTMV),
which is a single stranded DNA anellovius discovered only recently. However it is believed these
viruses replicate in the respiratory tract (Maggi, Pifferi et al. 2003), and one very recent report has
identified Betatorque virus in children with parapneumonic effusions and demonstrated that it is able
to infect and replicate within alveolar epithelial cells and induce innate immunity (Galmes, Li et al.
2012), potentially implicating it as a respiratory pathogen.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
192
Discussion
I have found similar bacterial species in BAL to those found in sputum, although the abundance was
higher in the latter sampling technique, suggesting a continuity of microbial communities between the
proximal and lower airways as well as the upper respiratory tract where expectorated sputum might
get contaminated during sample collection. I have not observed any general differences between
asthma and health in the composition of these microbial communities. However, I have identified two
individuals with high abundance of potential respiratory pathogens in BAL and it may be that future
larger studies will confirm these subjects to be representative of a subset of individuals with asthma
who have long term bacterial colonisation of the airways. Furthermore, these two individuals have
demonstrated proof of the concept that relationships can be identified within individual subjects
between airway microbial colonisation and local activation of the mucosal immune system. I have also
not found evidence of chronic persistent infection of the airway epithelium with respiratory viruses in
asthma.
Microbial modulation of the respiratory immune system
My aim in this analysis was to investigate the presence of potential microbial stimuli which might be
driving the immune response in asthma and to explore their relationship with the immunological data
that I generated in my cross-sectional study. Microbes might drive an aberrant immune response in
asthma in several ways: by providing ongoing stimulation of immune-pathology through local
activation of the mucosal immune system (Simpson, Grissell et al. 2007; Simpson, Powell et al. 2008;
Huang, Nelson et al. 2011), or by modulating the immune system systemically(Vael, Vanheirstraeten
et al. 2011), or they might play a significant role in the initial development of asthma. This latter
concept is sometimes termed the ‘hygiene hypothesis’(Strachan 2000): the suggestion that there is a
preventive effect of early childhood infections on the risk of allergic sensitisation, based on
relationships between risk of asthma and childhood family size (Strachan 1989; Strachan 1997;
Strachan 2000; Cullinan 2006), attendance at day care (von Mutius 2007), childhood bacterial
infections (von Mutius 2007), exposure to non-viable microbial products (Riedler, Braun-Fahrlander et
al. 2001) or to greater environmental microbial diversity (Ege, Mayer et al. 2011). These relationships
are particularly strong if the exposure occurs in the first year of life (Ege, Bieli et al. 2006; Loss, Bitter
et al. 2012). This may be related to the acquisition of different types of microbial flora early in life
(Bisgaard, Hermansen et al. 2007; Thavagnanam, Fleming et al. 2008; Roduit, Scholtens et al. 2009),
which in turn could be influenced by antibiotics (Wickens, Pearce et al. 1999; Droste, Wieringa et al.
2000; Noverr, Noggle et al. 2004; Russell, Gold et al. 2012).
Antibiotics may be modulating asthma risk by their effects on the faecal microbiome (Bisgaard, Li et
al. 2011; Vael, Vanheirstraeten et al. 2011; Han, Huang et al. 2012). Gastrointestinal (GI) flora may
play an important role in the induction of tolerance to airway allergens (Maeda, Noda et al. 2001;
Noverr, Noggle et al. 2004) mediated by Treg to down-regulate airway TH2 responses to the same
antigens (Noverr, Noggle et al. 2004). Experimental allergic airways disease is exacerbated in germ-
free mice compared with normal, and this exaggeration can be reversed by GI recolonisation with
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
193
normal commensal flora (Herbst, Sichelstiel et al. 2011). Commensal gut microbes can also modulate
the generation of virus-specific T cells (Ichinohe, Pang et al. 2011) or produce anti-inflammatory short
chain fatty acids by fermentation of dietary fibre (Maslowski, Vieira et al. 2009).
In summary microbes may play a key role in initiating asthma and also act later in life by driving
activation of the airway mucosal immune system either through the distant immunomodulatory effects
of the GI microbiome or more directly through the local presence of an airways microbiome.
Low bacterial frequencies argue against a significant airways microbiome
The key observation from my data is that in BAL from 47 subjects across a spectrum of health and
asthma I did not find evidence of a complex commensal airways microbiome in health or asthma. In
general the number of sequences (reads) detected was low and it was rare to find multiple sequences
from a single OTU in any given sample. These observations argue against the existence of a
consistent and distinct microbiome in the airways or healthy or asthmatic subjects. Of note, my study
included severe asthmatics in whom on might expect that microbes could be playing a more important
role, especially in neutrophilic forms.
This would contrast with the conclusions drawn of Hilty et al. (Hilty, Burke et al. 2010) who studied 11
adult asthmatics and eight healthy controls by sequencing DNA for the bacterial 16S RNA genes
found on protected airway brushings. These authors suggested that the ‘bronchial tree contains a
characteristic microbial flora that differs between health and disease’. The numbers in their study are
small, and the method was different, quantifying DNA by semi-quantitative PCR and sequencing DNA
after cloning in bacteria. It is not clear that these conclusions are valid from the data presented,
because they report cladistic analysis showing that bronchial microbial communities clustered with
oropharyngeal in health. This would rather support the conclusions of Charlson et al. that ‘bacterial
populations in the healthy lower respiratory tract (LRT) largely reflect upper respiratory tract (URT)
organisms, likely resulting from transient entry rather than independent communities with
indistinguishable structure (Charlson, Bittinger et al. 2011). Furthermore, unlike the work by Charlson
et al., the study by Hilty et al. did not present separate analysis of upper airway microbiota, specify the
route of intubation, nor include environmental controls. Hilty et al. report a finding of 2000 genomes
cm-2 of bronchial surface, but they do not present a separate analysis between asthma and health,
rather basing this figure on a mixed population comprising healthy controls, asthmatic individuals and
subjects with COPD. It is very unlikely that bacterial counts in COPD would be similar to those found
in health as chronic bacterial colonisation of COPD is well documented(Hill, Campbell et al. 2000;
Sethi, Evans et al. 2002; Wilkinson, Patel et al. 2003; Hurst, Wilkinson et al. 2005). It is therefore
misleading to present a global estimate of airway colonisation as a single figure from such a mixed
population. It is also inappropriate to present a mean value when the bacterial genome copy numbers
measured ranged from 62 to 210,000, suggesting a data-set which is highly skewed, most likely by
the inclusion of these heavily colonised COPD subjects.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
194
Charlson et al. present a very careful analysis of airway microflora from multiple sites within each
individual’s respiratory tract; nasopharyngeal swabs, oropharyngeal swabs, oral washes, two swabs
from the tip of the bronchoscopes, three bronchoalveolar lavage samples (Charlson, Bittinger et al.
2011). Furthermore upper airway contamination was minimised by the use of oral intubation and 2
sequential bronchoscopes, whilst environmental contamination was controlled by also analysing a
wash taken from each bronchoscope prior to the procedure. These results convincingly supported
their conclusions that in health ‘the lung does not contain a consistent distinct microbiome, but instead
contains low levels of bacterial sequences largely indistinguishable from upper airway flora’. As
mentioned above I would argue that a correct interpretation of the data of Hilty et al. would also
support this conclusion.
Unique contributions from this thesis
Can my data add to the work by Charlson et al.? Charlson examined only six healthy controls, and the
logical extension was to compare health with asthma, which I have done. However, my data have
several limitations. First, they did not include the environmental controls used by Charlson. This would
have conclusively detected the suspected contamination of the bronchoscopes. In future work I would
include these controls, but in the event the pattern and nature of the contamination could be clearly
deduced and do not detract from the key findings. Second, I have sampled only sputum and BAL.
However the work by Charlson et al. has now demonstrated that continuity of microbial communities
between different airway samples means that the same communities may be detected by BAL or by
brush or oral wash, and so the exact choice of sampling technique is not critical, and it is rational to
select a single endobronchial sampling technique. Third, I used only a single bronchoscope, and in
some individuals this was by nasal intubation. I have however used a different method of minimising
upper airway contamination: using a sterile double lumen BAL catheter which was sealed with a wax
plug and was introduced only after the bronchoscope lumen had in effect been washed with 10-16 ml
of lidocaine solution. This technique is probably comparable in sterility to the use of protected brushes
as they are not actually sealed with a wax plug, but as I have noted a small area of the bronchoscope
tip which forms the wedge will contact the BAL fluid.
Notwithstanding these limitations, my work also has several strengths, principally the use of whole-
genome sequencing. To date this technique has not been applied to airway samples in asthma and
has the unique advantage of also detecting organisms which do not express bacterial 16S RNA,
namely viruses and fungi. Furthermore, whole-genome sequencing can also identify entirely novel
pathogens, so is the method of choice for pathogen discovery, which is of particular use in detecting
novel viruses causing acute exacerbations of asthma.
My description of the subject with H.Influenzae infection provides proof of concept for the use of shot-
gun metagenomics in clinical care. However as Charlson has noted, unlike quantitative thresholds
empirically determined for diagnosis of pneumonia, there are no validated criteria for defining
colonization or identifying normal microbial populations of the lower airways(Charlson, Bittinger et al.
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
195
2011). Determination and validation of such thresholds will require large scale clinical studies using
highly standardised protocols and preferably linked prospectively to therapeutic treatment decisions.
The detection of T whipplei in one individual adds to the emerging evidence(Bousbia, Papazian et al.
2010; Charlson, Bittinger et al. 2011; Fenollar, Ponge et al. 2012; Urbanski, Rivereau et al. 2012) that
this poorly understood bacterium may be distinguished by an unusual ability to achieve long term
colonisation of the airways, perhaps only in immunologically susceptible individuals, and perhaps
functioning as an opportunistic pathogen. On the other hand, it might be argued that given the high
sensitivity of PCR methods, this could have been an incidental finding of no pathological significance.
PCR techniques have shown high rates of T whipplei carriage in asymptomatic individuals in saliva
and faeces (Ratnaike 2000; Rolain, Fenollar et al. 2007; Fenollar, Laouira et al. 2008; Fenollar, Trape
et al. 2009). If present in saliva, it is feasible that this bacterium might have arrived by microaspiration
of saliva. Nonetheless, with the tools now available there are increasing reports of T whipplei being
identified solely in the respiratory tract, either in patients with symptoms of pulmonary Whipple’s
(Fenollar, Ponge et al. 2012) or culture-negative pneumonia (Bousbia, Papazian et al. 2010). It is
intriguing to note that in the study by Charlson et al. (Charlson, Bittinger et al. 2011) using deep
sequencing in healthy individuals, of 3431 total OTU identified, there was only one which was present
at high abundance (32 sequences) in all lower respiratory tract samples and absent from all upper
respiratory tract samples of the same individual and this was T whipplei (Charlson, Bittinger et al.
2011). This was considered to represent a genuine detection of a bacterium, which supports the
hypothesis that the 35 sequences we detected also represent genuine presence of the bacterium.
There is certainly much that is unknown about T whipplei. It is interesting to note that as awareness of
the condition has increased over the last decade, due to the application of PCR techniques, and the
number of samples sent for testing has increased dramatically, there has been no change in the
positive ratio of tested samples (Edouard, Fenollar et al. 2012), which implies that T whipplei remains
a markedly under-diagnosed condition. It is plausible that the current intensive investigation of the
lung microbiome will identify the T whipplei as one of a small number of bacteria – along with the
phylogenetically related mycobacteria - capable of achieving long term colonisation of this unique
anatomical niche, perhaps only in immunologically susceptible individuals(Lagier, Fenollar et al.
2011).
No evidence of chronic respiratory viral infection in asthma
My failure to detect evidence of any particular virus at significant copy number in either BAL or
sputum from these cross sectional samples argues against the presence of chronic active viral
infection in the respiratory tract as a pathogenic mechanism in asthma.
The metagenome includes all organisms that live on us or in us. By extrapolation from seroprevalence
studies it is estimated that humans are chronically infected with 8-12 viruses, such as herpesviruses,
cytomegalovirus, anelloviruses, Epstein-Barr virus (EBV), and JC and BK polyomaviruses(Virgin,
Wherry et al. 2009). The high species specificity of polyomaviruses, which infect 72-98% of humans,
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
196
suggests a prolonged period of coevolution with humans. Viruses can maintain latent infection in
three ways: continuous replication (e.g. human immunodeficiency virus, hepatitis B and C), latency
and reactivation (e.g. EBV) and invasion of the genome (e.g. endogenous retroviral elements which
are transmitted vertically). Many viruses maintain latency by subversion of immunity which may have
a number of general effects on the immune system including persistent secretion of proinflammatory
cytokines and potential skewing of T cell response towards effector rather than memory
phenotypes(Virgin, Wherry et al. 2009). Of particular relevance to this thesis is the observation in a
murine model for respiratory syncytial virus (RSV) that prolonged presence of Sendai virus, is
associated with an iNKT mediated activation of IL-13 secretion from macrophages which contributes
to allergic airways disease (Kim, Battaile et al. 2008). This occurs even though the virus is cleared to
non-infectious trace levels by day 12 yet the IL-13 secreting macrophage phenotype does not develop
till day 21 post infection. Thus infections that are generally considered harmless or unimportant play a
role in shaping the normal immune response, at the cost of introducing immunopathology in
susceptible individuals (Virgin, Wherry et al. 2009).
Is there evidence of this occurring in human asthma outside the setting of an acute exacerbation? The
possibility for chronic persistence of human rhinovirus (RV) has been demonstrated in
immunosuppressed transplant recipients by Kaiser et al. who recurrently isolated viable RV of the
same strain from 2 lung transplant recipients over a 12 month period(Kaiser, Aubert et al. 2006). In
asthmatics Harju et al detected RV by PCR in sputum more often than in health and noted that
asthmatics positive for RV had worse symptoms and poorer lung function(Harju, Leinonen et al.
2006). Using immunohistochemistry Wos et al. found RV more in 9/14 (63%) of bronchial biopsies
from asthmatics but only 2/6 (33%) of controls, whilst using in situ PCR she found RV in 73% of
asthmatics and 22% of control biopsies (P<0.001) and again presence of RV correlated with poorer
lung function and worse eosinophilic inflammation(Wos, Sanak et al. 2008). Malmstrom et al.
detected RV by PCR in 45% of bronchial biopsies from infants with persistent wheeze and again
presence of RV correlated with worse lung function (Malmstrom, Pitkaranta et al. 2006). The
fundamental problem with all these studies is that RV replication persists much longer than, and also
precedes, the period of upper respiratory tract symptoms, and so it is very hard to determine that
detection of RV is not associated with a recent acute viral infection, or even one which is about to
occur. Jartti et al. found that 16% of asymptomatic healthy children were positive for RV, of which
38% developed symptoms in the subsequent week, and found that RV takes at least 5-6 weeks
(Jartti, Paul-Anttila et al. 2009) to become undetectable by PCR, perhaps longer(Kling, Donninger et
al. 2005). None of the studies in asthma mentioned above allowed for this: Wos sampled at least
three weeks after an exacerbation, Harju four weeks, and Malmstom had no period of quarantine, and
therefore noted higher rates of RV detection in subjects who had suffered a symptomatic URT
infection within the previous six weeks. Furthermore it is recognised that RV secretion persists for
longer in asthmatics (Corne, Marshall et al. 2002) due to defects in the induction of type I and III
interferons (Wark, Johnston et al. 2005; Contoli, Message et al. 2006).
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
197
I ensured that subjects had been free from symptoms of respiratory infection for at least six weeks
prior to sampling, which may explain why I found no evidence of chronic respiratory viral infection.
The most consistent interpretation of all these studies is that RV does not cause chronic infection in
asthmatic airways during periods of clinical stability, but rather is frequently detectable by sensitive
molecular methods several weeks after an overt or occult acute upper respiratory tract infection, and
this finding is more frequent in asthmatic people due to the well understood defect in antiviral innate
immunity.
Conclusion
In summary, I have presented a novel application of whole-genome shot gun sequencing to the
analysis of airway microbial samples. My data argue against the existence of a distinct airway
microbiome in health or in asthma, and support the conclusion that microbes within the lung are in
general a transient result of microaspiration of upper airway flora. Conversely in specific cases
chronic low grade infection with opportunistic pathogens or true pathogens may drive the
immunopathology of asthma, and perhaps this is particularly true in neutrophilic phenotypes. In a
single BAL sample I have directly shown the presence of an active infection with a true pathogen
inducing an exuberant BAL TH17 response, and in another individual a deficiency of BAL and
bronchial TH17 cells appears to be linked, perhaps causally, to the presence of an opportunistic
pathogen. Finally, I have found no evidence to support the hypothesis that chronic persistent viral
infection drives airways inflammation in asthma during periods of clinical stability, although my data
do lend weight to the emerging evidence that the recently described Betatorquevirus may be a
respiratory pathogen.
Future work
There would be much value in conducting future studies to further explore the relationship between
asthma and the human microbiome, particularly in addressing the following questions.
I. Whilst there is no evidence of a commensal airway microbiome in most individuals, are there
subsets of asthmatics in whom chronic colonisation with airway microbes occurs, and does
this respond to directed antibiotic therapy? This may be particularly relevant in patients who
have bronchiectasis associated with asthma. Patients with associated chronic rhinosinusitis
might also be a subgroup of asthma in whom microbes could play a more important role and
a study of the microbiome in these could provide more information than could be obtained by
simple microbial culture.
II. If such individuals exist, what is the mechanism by which they have become colonised: is this
attributable to genetic or epigenetic defects in innate immunity, or is it perhaps a result of the
immunopathology of chronic asthma causing impairment of mucosal immune function, or is it
a consequence of deficiency of adaptive immunity, such as MAIT cells? Is such a deficiency a
primary phenomenon, or is it secondary to therapy such as with corticosteroids?
III. What is the nature of the relationship between the gut microbiome and the airway mucosal
immune system in humans?
Timothy SC Hinks 6. Deep sequencing of the airway microbiome
198
IV. Can the use of whole genome sequencing be validated and developed for clinical application in
respiratory medicine?
Such studies need careful design. My data would suggest the importance of highly standardised
protocols for collection of specimens, the priority of collecting controls for environmental
contamination and the value in collecting simultaneous upper airway and stool samples for
metagenomics. Bronchoscopes used should be handled and cleaned in a dedicated facility which
does not process endoscopes used in the GI tract or other sources of high biomass contamination.
Airway sampling could be best achieved either by using wax plug protected brushes or by bronchial
lavage. In the latter case, samples should ideally be concentrated by ultracentrifugation to
compensate for the very low biomass present in the human airways.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
199
CHAPTER 7
T cell phenotypes during natural cold-induced
asthma exacerbations I see no reason to call it by its Greek name, difficulty in breathing being a perfectly good way
of describing it. Its onslaught is of very brief duration – like a squall, it is generally over within the
hour. One could hardly, after all, expect anyone to keep on drawing his last breath for long,
could one?...doctors call it a ‘rehearsal for death’, since eventually the breath does what it has
often been trying to do. 7
7 The Stoic philosopher Lucius Seneca’s (4 BC-AD 65) vivid description of his own symptoms, perhaps the earliest known personal description of asthma. Seneca, Epistulae Morales ad Lucilium, c.AD 62-5
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
200
Introduction
The data presented in earlier chapters of this thesis all concerned samples taken during periods
of clinical stability. However temporal variability of symptoms and lung function is a cardinal
feature of asthma (Hyde 1860; Bousquet, Jeffery et al. 2000) and this is manifest most clearly
during an acute exacerbation. As T cells are part of a complex immune system perpetually
responding to dynamic changes in antigenic stimulation I will present in this chapter the results
of a longitudinal investigation into the dynamics of CD4+ T cell responses during naturally
occurring acute exacerbations. I will first briefly define the concept of an exacerbation and
present a summary of what is known about the dynamics of the associated immune response,
particularly with respect to IL-17 and TH17 cells, the subject of this thesis and also IFN-β1α,
which was administered to some of the participants.
The nature of asthma exacerbations
The definition of an exacerbation is still a subject of discussion (Dougherty and Fahy 2009;
Fuhlbrigge, Peden et al. 2012), but acute exacerbations are defined in the GINA guidelines as
‘episodes of progressive increase in shortness of breath, cough, wheezing, or chest tightness,
or some combination of these symptoms, accompanied by decreases in expiratory airflow that
can be quantified by measurement of lung function’ or also as an ‘acute and severe loss of
control that requires urgent treatment’ ((GINA) 2010).
Whilst exacerbations can be triggered by a variety of factors including allergens, pollutants,
emotional stress and drugs ((GINA) 2010), the triggers in the majority of exacerbations are
acute viral infections of the upper respiratory tract (Johnston, Pattemore et al. 1995; Johnston,
Pattemore et al. 1996). Many asthmatic individuals suffer from increased and more severe lower
respiratory tract symptoms during these infections due to a defect in the production of type I
(Wark, Johnston et al. 2005) and type III (Contoli, Message et al. 2006) interferons. Viruses are
detected by PCR in approximately 80% of exacerbations (Johnston, Pattemore et al. 1995) and
are associated with airway neutrophilia (Wark, Johnston et al. 2002).
The immune response to rhinovirus
Approximately 2/3 of these viruses are identified as rhinovirus (RV)(Kelly and Busse 2008). RV
infection induces inflammation, and airway recruitment of neutrophils, eosinophils, mast cells,
CD4+ and CD8+ T cells, via increased IL-6, -8, -16, eotaxin, IFN-γ-inducible protein 10 (IP10,
CXCL5), and regulated and normal T cell expressed and secreted (RANTES, CCL5). Murine
models have shown that RV infection induces TH1 and TH2 cytokines, and exacerbates TH2
response to allergen challenge (Bartlett, Walton et al. 2008).
Regarding the dynamics of the T cell response to rhinovirus, in vitro analysis of human tonsillar
tissue shows that RV evokes a dose-dependent, CD4-dominant T cell response, with a peak of
IL-2 secretion at 24 hours and IFN-γ at 3 days (Wimalasundera, Katz et al. 1997). A study of
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
201
children with tracheostomies found virus-specific and bystander CD8+ cells migrated to the
lungs during acute respiratory viral infection, accompanied by a reciprocal fall in peripheral
antigen-specific T cells and transient increase in the CD8:4 ratio (Heidema, Rossen et al. 2008),
consistent with similar findings in mice (Levandowski, Ou et al. 1986).
TH17 cells may play an important role in the antiviral host response as they induce human β-
defensins, which are antiviral and can recruit memory T cells via CCR6 (Wolk, Kunz et al. 2004)
and may infiltrate bronchial mucosa during an asthma exacerbation (Pene, Chevalier et al.
2008). In healthy humans IL-17 modifies the responses of in vitro cultured epithelial cells to
rhinovirus, enhancing virally-induced synthesis of IL-8 and β-defensin and consequent
neutrophilic inflammation, whilst suppressing induction of the eosinophilic chemokine regulated
and normal T cell expressed and secreted (RANTES)(Wiehler and Proud 2007). In mice
pulmonary viral infection is associated with TH17 recruitment (Lochner, Peduto et al. 2008),
higher IL-17 expression and mucus hyper-secretion (Hashimoto, Graham et al. 2004;
Hashimoto, Durbin et al. 2005). However human data on the dynamics of anti-viral TH17
responses are lacking.
Whilst I have not observed differences in TH17 immunity during clinically stabile asthma, it was
necessary to investigate the possibility that asthma exacerbations might be associated with
aberrance in the dynamics of the TH17 response during acute exacerbations. In conjunction with
a controlled trial of inhaled recombinant human (rh)IFN-β1α for the prevention of asthma
exacerbations during the common cold I undertook longitudinal follow-up of a well characterised
cohort of asthmatics with frequent exacerbations with the aim of studying how TH17 cells
change during virus infections and associated asthma exacerbations. Furthermore, as in vitro
(Ramgolam, Sha et al. 2009; Wenink, Santegoets et al. 2009; Zhang, Jin et al. 2009) and
animal data (Guo, Chang et al. 2008; Martin-Saavedra, Gonzalez-Garcia et al. 2008; Orgun,
Mathis et al. 2008; Shinohara, Kim et al. 2008; Chen, Chen et al. 2009) have suggested IFN-β
may influence TH17 differentiation, I was able also to investigate how treatment with IFN-β1α
influences TH17 function in vivo.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
202
Study design
Interferon beta study
I undertook longitudinal follow-up of subjects with moderate asthma and a history of frequent
exacerbations who were participating in a phase II, double-blind, randomised, placebo-
controlled trial (“SG005”, NCT01126177 (SynairgenResearchLtd 2012)) of inhaled recombinant
human (rh)IFN-β1α given at the onset of a common cold to asthmatic patients with the aim of
preventing/ameliorating an exacerbations (Figure 2.2). Subjects were screened at baseline then
recalled for a second study visit within 24 hours of developing symptoms of an upper respiratory
tract infection. At this stage subjects were randomised to receive 6 MIU rhINF-β1α or placebo
once daily for 14 days via an I-neb Adaptive Aerosol Delivery device (Philips Respironics,
Guildford, UK)). Subjects returned for scheduled visits on days 4, 7, 10, 13, 17 and days 44-48
after the first onset of symptoms. At all visits subjects underwent clinical assessment and
phlebotomy. In addition sputum induction was performed on the baseline visit (V1) and visits 3
and 4 (days 4 and 7 respectively) (see study schedule, Table 7.1). Subjects also performed
home lung function monitoring and Asthma Index score (Sorkness, Gonzalez-Fernandez et al.
2008) reporting twice daily, Jackson Cold score (Jackson, Dowling et al. 1958) reporting once
daily, and shortened-Asthma Control Questionnaire (ACQ (Juniper, Svensson et al. 2005))
reporting weekly throughout the treatment phase.
I was an active member of the team conducting this study in the capacity of (honorary) clinical
research fellow along with 3 other clinical fellows and I also participated in the processing of the
PBMC along with other members of the laboratory team. I conducted all the processing of
sputum on the samples gifted to me and all processing of PBMC subsequent to the initial
isolation, or cryopreservation of PBMC (see acknowledgements).
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
203
Table 7.1 Study schedule for longitudinal study
Visit Number V1 V2 V3 V4 V5 V6 V7 V8
Pre-
treatment
phase
Treatment phase
Assessment days Screening
Day 1
(Within 24h
of cold
symptoms)
Day
4
Day
7
Day
10
Day
13
Day
17
Day 44-
48
Consent X
Medical history X
Physical examination X x x x x x x
Vital signs X x x x x x x
Height and weight X x
12 lead ECG X
Skin allergy test x
FENO x x x x x x x
FEV1, FVC, PEFR x x x x x x x
TLCO x x x
PD20 x
Home monitoring x x x x x x x
Urinalysis x x x
Nasal lavage x x x
Dose administration x x x x x
Phlebotomy
(cryopreserved) x x x x x x x x
Phlebotomy (fresh) x x x
Sputum induction (fresh) x x x
ECG, electrocardiogram; FENO, fractional exhaled nitric oxide; FEV1, forced expiratory volume
in 1 second; FVC, forced vital capacity; PEFR, peak expiratory flow rate; PD20, provocative
dose 20; TLCO, transfer factor carbon monoxide.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
204
Immunological samples
Two types of sample were available to me:
i) ‘fresh’ samples of paired peripheral blood and induced sputum at visits 1 (baseline), visits
3 (day 4) and visits 4 (day 7). Not all subjects successfully produced induced
sputum at each visit, and so these samples were only available on a subset of 31
subjects (see Figure 2.2). PBMC were only analysed freshly if there was an
adequate paired sputum sample from that visit. Adequate samples were obtained
from 14 subjects at V1, 13 subjects at V3 and 14 subjects at V4.
ii) cryopreserved samples of peripheral blood mononuclear cells from every subject at
every study visit. Complete paired sets of sample from all 8 study visits have been
analysed from 26 of these subjects, 13 who were randomised to rhIFN-β1α and 13
who were randomised to placebo.
As the process of cryopreservation significantly affected the phenotype of these cells as
measured by intracellular cytokine staining (ICS), immunological data from these two sets of
samples will be presented separately, in separate results sections.
In addition some PCR data were obtained from samples gifted to me from a pilot study called
‘SG007’. This was a single-group, unblinded rhinovirus challenge study in which 11 moderate
asthmatics were challenged with 100 tissue culture infectious dose 50 (TCID50)/mL) of human
RV16 (Parry, Busse et al. 2000; Adura 2013). Phlebotomy, sputum induction and nasal lavage
were performed at baseline and at 6 follow-up visits over the following 14 days (Adura 2013).
First I will present data from this pilot study SG007 (Results I), then I will present an analysis of
fresh samples from SG005 (Results II), then I will present data from the SG005 cryopreserved
samples (Results III).
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
205
Study populations
Pilot RV16 challenge study
Forty-four non-smoking, moderate asthmatics were screened for serological evidence of
humoral immunity to RV16. Eleven subjects had no humoral immunity to RV16 and received
viral inoculation. They comprised 4 male and 7 female subjects with a median age of 38 years
(range 20-53 years) with a median baseline FEV1 98.0% of predicted (IQR 85.5-105.5). All were
receiving ICS with a median dose of 400 mcg BDP equivalent (400-900). All 11 subjects
developed active infection, as evidenced by a) sero-conversion (>4-fold increase in anti-RV16
neutralising antibodies in serum during convalescence) (n=8/11), b) shedding of virus in the
nose detected by qPCR of nasal lavage (n=10/11) and c) virus detection by qPCR in the lower
respiratory tract as measured in sputum (n=6/11) (Adura 2013). The cDNA from these samples
was gifted to me and I had no part in prior collection or processing of the samples (see
acknowledgements).
Interferon-beta study longitudinal cohorts
The study was approved by the Southampton and South West Hampshire Research Ethics
committee A (REC number 10/H0502/14). Subjects were aged 18 to 65 with symptoms of
asthma for at least 2 years confirmed by medical history and ≥12% and 200mL bronchodilator
reversibility or evidence of BHR. Subjects were on maintenance ICS. Subjects had a history of
virus-induced exacerbations of asthma with at least one exacerbation in the last 24 months (but
not within the last 1 month) requiring oral steroids or antibiotics and answered ‘Yes’ to the
question ‘Does a cold make your asthma worse?’ Baseline clinical characteristics of those
subjects included in the analysis of fresh and of cryopreserved samples are shown in Table 7.2
and Table 7.3 respectively.
Additional criteria for entry to randomisation and the treatment phase included a history of
respiratory virus symptoms that had developed within the last 24 hours, defined as either:
Cold symptoms (specifically a blocked or runny nose, and a sore or scratchy throat) or
Influenza-like illness (Fever >37.8 °C plus two of the following: headache, cough, sore
throat and myalgia)
As shown in figure 2.2, 120 subjects were assessed for eligibility and consented at
Southampton, of which 102 successfully completed the baseline screening visit and satisfied all
inclusion and exclusion criteria. During the period of the study (March 2010 - December 2011)
47 subjects developed symptoms of an acute URTI and were randomised. Of these fresh
samples were obtained from 31 subjects (Table 7.2) and complete series of paired
cryopreserved samples were obtained from 26 subjects (Table 7.3).
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
206
Table 7.2 Clinical characteristics of the longitudinal cohort (fresh samples)
n 31DemographicsSex (M/F) 12 / 19Age (median [range], years) 30 (19-63)Pulmonary function
FEV1 (% predicted) 106 (89-112)FEV1 reversibility (%) 5.7 (1.9-8.7)PEFR (% predicted) 101 (84-108)PEFR reversibility (%) 4.6 (2.5-11)PD20 (mg methacholine) 0.13 (0.082-0.45)
Exhaled nitric oxide (ppb, at 50 L/s) 22 (15-34)ClinicalAtopy (Skin prick positive, Y/N) 18 / 13
Peripheral eosinophil count (109/L) 0.2 (0.1-0.4)
Body mass index (kg/m2) 26.8 (24.4-30.5)Smoking status
Never / Former / CurrentDuration of asthma (years) 23 (19-28)ACQ score 0.71 (0.50-1.4)GINA level of control (n, %)
Controlled 0 (3.8)Partly controlled 27 (87)Uncontrolled 4 (13)
TreatmentInhaled steroids Yes
Dose (equivalent mcg BDP) 400 (400-900)Maintenance oral steroids (Y,N) NoLong acting β agonist (Y/N) 17 / 14Leukotriene receptor antagonist (Y/N) 1 / 30Antihistamine 6 / 25Step on BTS treatment algorithm 2 - 4
Exacerbation history (last 2 years)Courses of antibiotics 2 (1-4)Numer of exacerbatoins 4 (2-7)Hospital admissions 0 (0-0)Courses of oral steroids 2 (1-2)
Relevant comorbidities (n, %)Allergic rhinitis 20 (65)Eczema 7 (65)
ACQ, asthma control questionnaire; BDP, beclometasone dipropionate; BTS, British Thoracic Society; CT, computed
Values are medians with interquartile ranges, unless stated otherwise. N/A: not available.
All subjects
25 / 5 / 1
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
207
Table 7.3 Clinical characteristics of the longitudinal cohort (cryopreserved)
n 26 13 13DemographicsSex (M/F) 4 / 22 4 / 9 0 / 13Age (median [range], years) 30 (19-58) 30 (22-55) 27 (19-58)Pulmonary function
FEV1 (% predicted) 103 (87-111) 109 (86-118) 96 (89-106)FEV1 reversibility (%) 6.5 (2.9-8.8) 5.7 (2.8-8.9) 7.7 (4.0-8.6)PEFR (% predicted) 102 (85-109) 108 (84-120) 100 (87-107)PEFR reversibility (%) 4.2 (0-5.5) 3.0 (0.0-4.3) 4.8 (2.9-11)PD20 (mg methacholine) 0.15 (0.046-0.26) 0.24 (0.12-0.36) 0.09 (0.032-0.15)
Exhaled nitric oxide (ppb, at 50 L/s) 23 (15-37) 29 (17-41) 18 (12-26)ClinicalAtopy (Skin prick positive, Y/N) 16 / 10 8 / 5 8 / 5
Peripheral eosinophil count (109/L) 0.2 (0.1-0.3) 0.2 (0.1-0.3) 0.2 (0.1-0.3)
Body mass index (kg/m2) 27.5 (24.4-32.5) 25.1 (24.3-29.9) 31.2 (27.4-34.4)Smoking status
Never / Former / CurrentDuration of asthma (years) 20 (15-26) 20 (12-23) 23 (15-29)ACQ score 0.86 (0.43-1.3) 0.86 (0.43-1.3) 0.9 (0.57-1.57)GINA level of control (n, %)
Controlled 1 (3.8) 0 (0) 1 (7.7)Partly controlled 21 (81) 12 (92) 9 (69)Uncontrolled 4 (15) 1 (7.7 3 (23)
TreatmentInhaled steroids Yes Yes Yes
Dose (equivalent mcg BDP) 400 (400-800) 400 (200-400) 800 (400-800)Maintenance oral steroids (Y,N) No No NoLong acting β agonist (Y/N) 16 / 10 9 / 4 7 / 6Leukotriene receptor antagonist (Y/N) 1 / 25 0 / 13 1 / 12Antihistamine 6 / 20 4 / 9 2 / 11Step on BTS treatment algorithm 2 - 4 2 - 4 2 - 4
Exacerbation history (last 2 years)Courses of antibiotics 2 (1-4) 2 (1-4) 1 (1-3)Numer of exacerbatoins 5 (2-6) 5 (3-7) 3 (2-6)Hospital admissions 0 (0-0) 0 (0-1) 0 (0-0)Courses of oral steroids 2 (0-3) 2 (0-4) 2 (1-2)
Relevant comorbidities (n, %)Allergic rhinitis 18 (69) 8 (62) 10 (77)Eczema 8 (31) 3 (23) 5 (38)
Values are medians with interquartile ranges, unless stated otherwise. N/A: not available.ACQ, asthma control questionnaire; BDP, beclometasone dipropionate; BTS, British Thoracic Society; CT, computed tomogram; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; GINA, Global Initiative for Asthma; PEFR, peak expiratory flow rate; PD20, provocative dose 20.
All subjects Active rhIFN-β1α Placebo
21 / 4 / 1 10 / 3 / 0 11 / 1 / 1
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
208
Results I Analysis of pilot data from RV challenge cohort
Induction of IL-17 mRNA in sputum during experimental RV infection
In a pilot experiment to determine whether IL-17 was induced during acute respiratory viral
infections and if so what the magnitude of this induction was, I initially measured IL-17A mRNA
in whole sputum obtained from the SG007 pilot RV16 challenge study. Of 44 subjects who were
screened for inclusion in the study, 11 were challenged with virus and developed evidence of
infection. From these subjects only 4 provided usable sputum samples at more than one time-
point. Acute infection was associated with only modest induction of IL-17A mRNA (Figure 7.1).
The mean maximum fold-induction relative to baseline was 3.7, which occurred on day 3 post-
inoculation. By comparison there was a much greater, 2-3 log fold-induction of IFN-β in the
same subjects, with 1-2 log fold increases in other antiviral genes including interferon gamma-
induced protein 10 (IP-10) and myxoma resistance gene A (MxA) (Adura 2013).
Figure 7.1 Sputum IL-17 mRNA during experimental RV infection
IL-17 mRNA measured in whole sputum by RT-qPCR in 4 subjects undergoing experimental
infection with 100 TCID50 of rhinovirus 16 as part of clinical study SG007. Samples were
normalised to the housekeeping genes UBC and GAPDH. Graphs show means ± SD.
These preliminary data challenged the hypothesis that viral infections would produce a major
induction of IL-17A. It is possible that this was partly due to the nature of RV16 which is known
to produce a fairly mild clinical syndrome (Fleming, Little et al. 1999; Grunberg, Timmers et al.
1999). Nonetheless, all subjects did experience a significant increase in URTI symptoms (mean
increase in Jackson Cold Score of 8.6, P<0.001) and in asthma symptoms (P<0.001), with
concomitant falls in FEV1 and PEFR (Adura 2013). Furthermore they provided data needed to
inform sample size calculations for a proposed viral challenge study, which ultimately was not
considered feasible.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
209
Results II Analysis of fresh samples from longitudinal cohort
Next I analysed data obtained from ‘fresh’ samples processed immediately ex vivo from the
longitudinal study of naturally-occurring viral infections.
T cell frequencies in peripheral blood and sputum during acute viral infection
Frequencies of T cells in peripheral blood and sputum from all 31 subjects combined are
presented in Figure 7.2. Unfortunately very few subjects produced usable samples from sputum
induction at the baseline visit (n=14 samples out of 102 attempted inductions, Table 7.4) and
these were therefore not well paired with the samples obtained at symptom days 4 and 7, so
paired statistical tests could not be used and sample sizes overall were small (n=13-14 at each
visit). Therefore no statistically significant differences were observed in T cell frequencies over
time (ANOVA P>0.15), except for TH17 cell frequencies in sputum, where differences of
borderline statistical significance were observed. Mean frequencies of sputum TH17 cells
increased 1.8-fold from 6.2% at baseline to 11% at symptom day 7 (ANOVA P=0.087, post-hoc
t test P=0.034).
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
210
Figure 7.2 T cell frequencies in peripheral blood and sputum during acute viral infection
Frequencies of T cells in peripheral blood and induced sputum during acute upper respiratory
tract infections measured by intracellular cytokine staining and flow cytometry on samples which
had not been cryopreserved. Day 0, baseline screening visit (n=14); day 4, symptom day 4
(n=13); day 7, symptom day 7 (n=14). Plots show means±95% confidence intervals (CI). Most
data are not paired. No differences were significant by ANOVA for any comparison.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
211
Table 7.4 Rates of successful sputum inductions during longitudinal study
Study visit
Number of subjects attending visit
Number of successful sputum inductions
Rate of successful sputum induction (%)
1. Screening 102 14 14
3. Symptom day 4 47 13 28
4. Symptom day 7 47 14 30
I had hypothesised that I would observe a dramatic early increase in sputum TH17 frequencies,
followed later by an increase in TH1 cells, mirrored by reciprocal falls in peripheral blood TH17
and TH1 frequencies. It is not possible to draw firm conclusions from this data-set, but whilst
there is evidence of the increase in sputum TH17 cell frequencies which I had expected, it is
unlikely the magnitude of such an increase would much exceed a doubling in TH17 cell
frequencies at most.
The effect of IFN-β1α on T cell frequencies in blood and sputum
Next I stratified these data from fresh samples according to allocation to IFN-β1α or placebo
(Figure 7.3). Statistical analysis was not possible because lack of pairing of data prevented
comparison of areas under the curve (AUC) between active and placebo, the data are not
corrected for baseline differences or adjusted for covariates and samples sizes were small.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
212
Figure 7.3 T cell frequencies in peripheral blood and sputum stratified by treatment
group
Data presented in Figure 7.2 stratified by whether subjects received inhaled active rhIFNβ1α
(continuous lines n=4-6) or placebo (broken lines n=10-8). Samples had not been
cryopreserved. Plots show means ± 95% CI. Most data are not paired. No significant differences
were observed between groups.
Results III Analysis of cryopreserved PBMC samples from
longitudinal cohort
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
213
In addition to these samples I also obtained much larger numbers of cryopreserved PBMC at 8
different time-points which provided a more extensive characterisation of the dynamics of the T
cell responses during acute exacerbations, and allowing more informative statistical analyses.
Frequencies of TH17, TH1, TH2 and Treg cells in PBMC are presented in Figure 7.4 for 26
subjects at 8 visits. T cell frequencies did not differ significantly over time by ANOVA in any T
cell subset. There was considerable inter-individual variation in T cell frequencies, which is
apparent when frequencies at multiple time-points a plotted separately for each subject Figure
7.5.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
214
Figure 7.4 T cell frequencies in cryopreserved peripheral blood during an acute viral
infection
Frequencies of major T cell subsets in peripheral blood during an acute upper respiratory tract
infection. T cells which were cryopreserved at baseline and 7 time-points from the onset of
symptoms were enumerated by intracellular cytokine staining and FACS. Plots show mean and
95% confidence intervals for n=26 individuals. Day 0, screening visit. Day 1, visit occurring
within 24 hours of developing first upper respiratory tract symptoms. Other visits occurred on
days 4, 7, 10, 13, 17 (each ±1 day) and day 30-35. (A) TH17 cells, (B) TH1 cells, (C) TH2 cells,
(D) Treg cells as % of total CD4+ T cells. (E) ratio of TH17 to Treg cells.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
215
Figure 7.5 T cell frequencies in cryopreserved peripheral blood during an acute viral
infection: showing individual subjects separately
The same data as shown in Figure 7.4 but plotted to show results from individual subjects
separately. (A) TH17 cells, (B) TH1 cells, (C) TH2 cells, (D) Treg cells as % of total CD4+ T cells.
(E) ratio of TH17 to Treg cells.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
216
Figure 7.6 Peripheral blood T cell subsets according to treatment group
A. Peripheral blood T cell frequencies plotted over time and stratified by whether subjects
received inhaled active rhIFNβ1α (n=13, continuous lines) or placebo (n=13, broken lines). Plots
show mean ±95% confidence interval. (A) TH17 cells, (B) TH1 cells, (C) TH2 cells, (D) Treg cells
as % of total CD4+ T cells. (E) ratio of TH17 to Treg cells. Differences in the areas under the
curves for treatment groups were compared by t tests, and were significant for TH17 cells
(P=0.006) and for the TH17/Treg ratio (P=0.006) only.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
217
Figure 7.7 Peripheral blood TH17 response according to treatment group
A. Peripheral blood TH17 cell frequencies measured by FACS plotted over time and stratified by
whether subjects received inhaled active rhIFNβ1α (n=13, continuous lines) or placebo (n=13,
broken lines). Plots show mean ±95% confidence interval.
B.Plot of areas under the curve for peripheral TH17 response over time, stratified by treatment
group. P=0.006 for unpaired t test.
TH17 cell frequencies in peripheral blood are elevated during treatment with inhaled
rhIFN-β1α
When subjects are stratified according to study randomisation, differences emerge between
those allocated to active IFN-β1α (n=13) and placebo (n=13) (Figure 7.6). Both the TH17 cell
frequencies and the ratio of TH17:Treg were significantly higher in subjects receiving active
treatment than placebo (P=0.006 for AUC comparison (Matthews, Altman et al. 1990)). The
comparison of the dynamics of the TH17 response is shown in more detail in Figure 7.7 where
the individual AUC for each subject are presented (Figure 7.7 B). The mean AUC was 2.2 fold
greater in subjects receiving IFN-β1α. As TH1, TH2 and Treg frequencies did not differ
significantly between treatment groups the change in the TH17:Treg ratio is attributable to
differences in TH17 cells alone.
TH17 cell frequencies in peripheral blood are according to whether subjects suffer an
asthma exacerbation
Could these IFN-β1α-induced differences in TH17 frequencies be due to drug treatment
preventing exacerbations? To address this possibility I stratified subjects instead according to
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
218
whether or not they developed an acute exacerbation of their asthma defined as a 0.5 point fall
in ACQ between screening and symptom day 7 (n=16 exacerbated, n=10 did not). The mean
magnitude of the AUC was 1.6 fold greater in subjects who exacerbated, but this did not reach
statistical significance (P=0.2 for AUC of TH17 response and P=0.12 for AUC of TH17:Treg
ratio)(Figure 7.8).
Figure 7.8 Peripheral blood TH17 response according to whether exacerbated
A. Peripheral blood TH17 cell frequencies measured by FACS plotted over time and stratified by
whether subjects experienced an exacerbation of their asthma defined as a 0.5 point fall in ACQ
between screening and symptom day 7 (n=16 exacerbated, n=10 did not). Plots show mean
±95% confidence interval.
B.Plot of areas under the curve for peripheral TH17 response over time, stratified by whether
experienced an exacerbation. P=0.2 for unpaired t test.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
219
Discussion
These longitudinal investigations into the dynamics of the immune response were undertaken
with the aim of determining whether IL-17 and TH17 cells are induced during naturally occurring
asthma exacerbations, and elucidating how treatment with a type-I interferon influences TH17
function. Two main conclusions can be drawn from the data presented in this chapter with
respect to these aims and the challenges of investigating airway T cell response in vivo in
humans.
Respiratory virus infections are not associated with a TH17 response
I hypothesised that airway accumulation of TH17 cells would occur early in infection, leading to
neutrophilia, followed by a TH1 dominant response. TH17 cells are induced early in immune
responses such as in early transplant rejection (Loeuillet, Martinon et al. 2006) or immunity to
mycobacterial infections (Khader, Bell et al. 2007; Umemura, Yahagi et al. 2007; Khader and
Cooper 2008) where they are present 3 days sooner than TH1 cells (Khader, Bell et al. 2007).
Indeed IL-17 can be produced even earlier by several innate-like cells such as iNKT cells
(Rachitskaya, Hansen et al. 2008), MAIT cells (Dusseaux, Martin et al. 2011) and γδ T cells
(Umemura, Yahagi et al. 2007; Khader and Cooper 2008). As the immune response matures
and subsequently polarises into TH1 or TH2 cells these can then rapidly suppress the more
transient TH17 response (Khader and Cooper 2008). Furthermore TH17 cells have been
implicated in viral exacerbations because of their ability to recruit neutrophils (Linden 2001;
Hellings, Kasran et al. 2003; Prause, Bozinovski et al. 2004; Oda, Canelos et al. 2005; Wiehler
and Proud 2007), which are the predominant cell type in the airways during exacerbations
(Message and Johnston 2001).
However, despite small numbers, my data do not show an early induction of a TH17 response.
In contrast to the 10-1000 fold induction of other anti-viral genes, I observed a mean maximum
3.7 fold induction of IL-17 in the RV16 challenge study. Whilst this study was small, challenge
studies have the particular strengths of a well-defined pathogen and precise knowledge of the
temporal course of the infection.
Similarly data from the study of naturally occurring exacerbations did not produce strong
evidence of major changes in the TH17 cell frequencies in peripheral blood in the cohort as a
whole. Although TH17 frequencies did increase in sputum, this change was of borderline
statistical significance (ANOVA P=0.087, post-hoc t test P=0.034) and modest magnitude (1.8-
fold). In addition IL-17 was not detectable at baseline or symptom day 4 in the serum from 58
asthmatics on BTS treatment step 4 and 5 in the SG005 study using a Luminex based assay
(Multi-Analyte Profiling, Myriad RBM, Austin, TX, USA)(Monk 2012). Furthermore there were no
significant differences in TH17 responses between subjects who did and who did not develop an
exacerbation. I did not include a group of healthy controls for comparison with asthma, but
arguably the comparison between asthmatics who did or did not exacerbate is as informative.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
220
Subjects who developed exacerbations in this study tended to be those who had more severe
asthma consistent with previous studies which have shown that patients with poorer baseline
control are more likely to experience exacerbations (Bateman, Bousquet et al. 2008). However it
is interesting that the mean baseline TH17 frequencies were identical amongst subjects who did
or did not exacerbate (Figure 7.8) implying that baseline TH17 cell frequencies are not a factor
determining the risk of an exacerbation. Thus together these data challenge the hypothesis that
asthma exacerbations might be characterised by a major dysregulation of IL-17 immune
responses.
Perhaps this lack of a strong antiviral TH17 response is because the predominant role of IL-17 is
immunity against bacteria and fungi. A deficiency of TH17 humans with heterozygous mutations
in STAT3 is associated with increased susceptibility to Staphylococcus aureus, Candida
albicans and bacterial pneumonias (Ma, Chew et al. 2008; Milner, Brenchley et al. 2008;
Woellner, Gertz et al. 2010) rather than viral infections. Whilst IL-17 can induce molecules like
β-defensin which has antiviral as well as antibacterial functions (Wolk, Kunz et al. 2004), it is not
known how important TH17 cells are for viral immunity in vivo. Indeed in an animal model TH17
cells promoted persistent of a viral infection by inhibiting virus-induced apoptosis (Grajewski,
Hansen et al. 2008). Probably the greater relevance of TH17 cells to respiratory viruses is that
virus induced suppression of IL-17 mediated by type I IFN may contribute to susceptibility to
secondary bacterial and fungal infections, as has been demonstrated in a murine model of
influenza infection (Kudva, Scheller et al. 2011).
Administration of inhaled rhIFN-β1α is associated with increased TH17 frequencies in
peripheral blood
It is because of this link between type I IFNs and IL-17 that the comparison of TH17 responses
between active- and placebo-treated groups is of interest. I hypothesised that administration of
inhaled rhIFN-β1α would inhibit the magnitude of the TH17 response to viral infection measured
in PBMC and airway samples. Contrary to this hypothesis I observed the opposite: evidence of
an increased TH17 cell response in peripheral blood. In the context of the potential therapeutic
use of rhIFN-β1α to prevent virus induced exacerbations of asthma, this is a reassuring finding.
Respiratory virus infections predispose to bacterial super-infections (Morens, Taubenberger et
al. 2008; Weeks-Gorospe, Hurtig et al. 2012) and in the case of rhinovirus this may be through
both disruption of epithelial barrier functions (Sajjan, Wang et al. 2008) and inhibition of T cell
function (Gern, Joseph et al. 1996). Moreover asthmatics are at particularly high risk of these
effects as severe asthma is a risk factor for invasive pneumococcal disease (Talbot, Hartert et
al. 2005; Klemets, Lyytikainen et al. 2010) and inhaled steroids are associated with increased
risk of pneumonia in subjects with airways disease (Calverley, Anderson et al. 2007; Crim,
Calverley et al. 2009; Welsh, Cates et al. 2010). Therefore it would be a concern if rhIFN-β1α
specifically impaired TH17 cell responses.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
221
Hypothetically TH17 cells might be suppressed by IFN-β. In animal models type I IFN favours
TH1 (Orgun, Mathis et al. 2008; Shinohara, Kim et al. 2008) or TH2 (Martin-Saavedra, Gonzalez-
Garcia et al. 2008) differentiation over TH17 and defects in the type I IFN receptor (IFNAR) lead
to increased IL-17 levels (Guo, Chang et al. 2008). There is less known in humans, although
there are some data related to the use of rhIFN-β1α intravenously for treating multiple sclerosis
(MS). In vitro IFN-β (Durelli, Conti et al. 2009; Ramgolam, Sha et al. 2009) or supernatant from
IFN-β-treated dendritic cells (Ramgolam, Sha et al. 2009) decreases TH17 frequencies in
PBMC, as well as decreasing RORC and IL-17A gene expression (Zhang, Jin et al. 2009).
However the situation may be different in vivo. IFN-β can increase the survival of CD4 cells (van
Boxel-Dezaire, Zula et al. 2010) and longitudinal follow up of 36 patients with MS found that
treatment with IFN-β was associated with decreased mRNA for IFNγ and T-bet but no fall in IL-
17 or RORC in peripheral blood (Drulovic, Savic et al. 2009). Given the difficulties of
extrapolating to humans data from in vitro analysis of PBMC or from animal models due weight
should be given to my data, which are the first to compare the effect on TH17 cells of IFN-β
compared with placebo in vivo in humans.
Timothy SC Hinks 7 T cell phenotypes during natural asthma exacerbations
222
Conclusion
In summary I have supplemented my data from my cross-sectional analyses of T cell subsets
with data on the dynamics of TH17 cell frequencies, showing that virus-induced exacerbations of
asthma are not associated with major fluctuations in TH17 cell frequencies. Furthermore
amongst subjects with URTIs I did not observe differences of significant magnitude in either the
baseline TH17 frequencies or the dynamics of the TH17 response between subjects who
developed an asthma exacerbation and those which did not. The use of rhIFN-β1α is not
associated with a suppression of TH17 cell immunity, but rather with a potentially beneficial
increase in peripheral TH17 cell frequencies.
These studies have highlighted the considerable challenges to investigation of the dynamics of
the airway immune responses in vivo. The rates of successful sputum inductions were low in
both studies, consistent with other asthma studies (Boniface, Koscher et al. 2003; Yoshida,
Watson et al. 2005; Papadopouli, Tzanakis et al. 2006; Mamessier, Lorec et al. 2007;
Mamessier, Milhe et al. 2007; Mamessier, Nieves et al. 2008) This is particularly problematic for
the study of rare cell populations and for intracellular cytokine staining which induces high rates
of apoptosis. These limitations constitute significant obstacles for longitudinal studies which are
especially sensitive to missing data (Matthews, Altman et al. 1990). Finally an additional
obstacle to the conduct of challenge studies with rhinovirus is the high prevalence of pre-
existing, cross-reactive humoral immunity.
These findings support the conclusions from my cross-sectional study that IL-17 and TH17 cells
do not play a significant role in the pathogenesis of human asthma.
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
223
CHAPTER 8
Discussion I was merely thinking God's thoughts after him. Since we astronomers are priests of the highest God
in regard to the book of nature, it benefits us to be thoughtful, not of the glory of our minds, but rather,
above all else, of the glory of God.8
8 Johannes Kepler (1571-1630)
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
224
My primary aim was detailed investigation of T cell phenotypes in asthma in relation to severity and
virus-induced exacerbations, with a particular focus on interleukin-17 TH17 cells and MAIT cells. My
intent was to translate advances in basic science and animal models into humans in vivo and to
improve characterisation of severe asthma versus milder forms of asthma, thereby facilitating future
progress in basic and applied research. In this final chapter I will summarise the findings of this
present work and their implications for our understanding of asthma in a wider context. I will then
discuss their implications for future research, some of which is in progress already.
The fundamental role of TH2 inflammation in asthma
Recent decades have witnessed a rapid expansion in our understanding of the variety of different
innate and adaptive T cell subsets (Shevach 2006; Lloyd and Hessel 2010), yet the cross-sectional
data presented in chapter 3 have again highlighted the pre-eminent role of TH2 cells in asthma, which
remains unchallenged 20 years after their initial recognition (Robinson, Hamid et al. 1992). In
extending such investigations to a wider spectrum of asthma phenotypes, my work has revealed
diversity in the patterns of T cell responses underlying distinct endotypes, such as the lower
peripheral blood TH2 bias in non-atopic asthma, or the deficiencies in BAL Treg, or sputum and biopsy
MAIT cells in more severe asthma. The study of TH2 cells in asthma is therefore likely to continue to
prove fruitful. Future work should focus on the mechanisms which can influence the persistence of a
TH2 response, such as epigenetic effects (Vijayanand, Seumois et al. 2012) which are potentially
amenable to pharmacological modulation or, immunotherapy (Robinson, Larche et al. 2004; Larche
2007; Roncarolo and Battaglia 2007).
The history of interleukin-17 and TH17 cells in asthma highlights research pitfalls
Using a number of techniques in a wide range of subjects and clinical samples I found little evidence
to support the now widely hypothesised role of IL-17 in asthma, nor have I found evidence that TH17
cells could be central players in chronic disease. These findings will, I hope, focus the attention of
other researchers away from this particular avenue. My data do suggest that IL-17 may be elevated in
a subset of mild, steroid naïve asthmatics, who suffer from allergic rhinitis and perhaps represent a
distinct endotype. My data also do not negate the findings by others that IL-17 mRNA or protein may
be raised in various samples of the asthmatic airways, so any further investigation of IL-17 should
address this specific phenotype and should focus on epithelial cells or eosinophils within the
respiratory mucosa as these are the most likely sources of the cytokine (Chakir, Shannon et al. 2003;
Doe, Bafadhel et al. 2010; Vazquez-Tello, Semlali et al. 2010; Howarth 2012; Jayasekera 2013).
IL-17 is frequently considered as a mediator of immune pathology, because of its importance in
inducing pro-inflammatory cytokines (Fossiez, Djossou et al. 1996) and in recruiting neutrophils
(Sergejeva, Ivanov et al. 2005; Fujiwara, Hirose et al. 2007; McKinley, Alcorn et al. 2008). However
this concept is an oversimplification as IL-17 has also been shown in animal studies to have
protective roles, functioning as a negative regulator of established inflammation (Schnyder-Candrian,
Togbe et al. 2006; Braun, Ferrick et al. 2008; O'Connor, Kamanaka et al. 2009; Murdoch and Lloyd
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
225
2010). Therefore future studies should include investigation of the dynamics of IL-17 in immune
responses such as acute allergen challenge and also investigation of the functional effects of IL-17 in
vivo perhaps by correlating with down-stream effects on immunological networks, such as by analysis
of the transcriptome of epithelial cells and other effector cells.
The negative findings from my studies of TH17 and γδ-17 cells constitute a refutation of what is at
present a popular hypothesis. This eventual failure of the TH17 hypothesis raises some general issues
and in particular underscores two pitfalls in asthma research. Firstly there is a danger in pursuing
hypotheses founded on clinical data based on animal models and relatively small human studies. As I
have outlined in the discussion of chapter 3 more than a decade of basic science and animal research
has been founded on evidence from just three clinical studies which have together been cited over
150 times in the literature (Molet, Hamid et al. 2001; Barczyk, Pierzchala et al. 2003; Chakir, Shannon
et al. 2003). These studies were of small size, with two enrolling only 6-10 asthmatics and the other
only including 6 healthy controls. They did not use the best techniques available at the time - such as
multi-colour flow cytometry (Krug, Madden et al. 1996) – nor robust statistical analysis (Barczyk,
Pierzchala et al. 2003) and crucially the data were never subsequently confirmed by other
investigators or even by data from the same groups.
The absence of robust human data has led in this case to another pitfall, that of overreliance on
animal models (Lloyd and Hessel 2010; Holmes, Solari et al. 2011) (Schnyder-Candrian, Togbe et al.
2006; Wakashin, Hirose et al. 2008; Wilson, Whitehead et al. 2009; Lloyd and Hessel 2010; Murdoch
and Lloyd 2010). Whilst these animal studies have been conducted to high standards and provided
fascinating insights into general T cell biology it can often be very difficult to extrapolate their findings
into a complex and uniquely human disease such as asthma. No animals, except perhaps cats or
horse, are known to suffer from asthma (Holmes, Solari et al. 2011) and even within a species there
are significant differences in outcomes between different strains and different immunisation protocols.
These protocols depend on unphysiological sensitisation procedures such as intraperitoneal injection
of high dose allergen in the presence of adjuvants (Zosky and Sly 2007), which are very unlike the
natural history of human asthma. Furthermore the allergic airways inflammation which ensues is
arguably more reminiscent of allergic alveolitis than asthma (Zosky and Sly 2007). Finally, few models
take account of the agents which usually trigger asthma exacerbations in humans such as infections,
air pollution, diet, tobacco smoke, drugs and other chemicals (Holmes, Solari et al. 2011).
An alternative approach is the development of ex vivo human models of airway immunology. Such
models are becoming more complex with the development of air-liquid interphase cultures of primary
bronchial epithelial cells (Swindle, Collins et al. 2009) which can retain important genetic and
epigenetic features of the human source and which are currently being developed to include
interactions with stromal cells and matrix, as well as the use of micro-fluidics to emulate the dynamics
of inflammatory cell influx (Swindle and Davies 2011) which will potentially enable modelling of
dynamic T cell-epithelial cell interactions. An alternative to this synthetic approach to modelling is the
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
226
use of whole tissue explant cultures which have the advantage of maintaining all the tissues cell types
of the mucosa in their entirety without destroying their functional networks (Nicholas, Staples et al.
2013). These models provide a potential platform for exploring truly integrated systems biology.
It is, also, likely that the application of a systems biology approach to assess whole interacting
networks of cytokines and inflammatory cells is going to be necessary to advance our understanding
of complex diseases such as asthma beyond what can be understood from a conventional
reductionist study of individual cells or cytokines (Sabroe, Parker et al. 2007; Cookson and Moffatt
2011; Zhang, Moffatt et al. 2012).
In conclusion, further research into the mechanisms, aetiology and clinical phenotypes of asthma
must always be driven by observations arising first from high quality, large scale studies in humans,
supplemented with novel, disease relevant ex vivo human models and the application of systems
biology.
A renewed interest in CD8+ T cells in asthma is warranted
Data presented in chapter 4 demonstrated an important relationship between CD8+ T cells and
asthma, and are a timely reminder that these somewhat neglected cells should merit further research.
Such investigation should focus on the specific clinical endotypes in which they are implicated by my
data: subjects with eosinophilic asthma and a history of nasal polyps and smoking. The relationship
with smoking and nasal polyps suggests potential mechanisms which might lead to the development
of Tc2 inflammation, such as smoking-related oxidative stress (Pierrou, Broberg et al. 2007), or nasal
colonisation with pathogenic bacteria such as S.aureus which produces staphylococcal enterotoxins
associated with allergic rhinitis, nasal polyps and asthma (Bachert, Gevaert et al. 2007). An important
issue to research is the nature of the antigen specificity of these CD8+ cells to determine whether
their primary specificity is to respiratory, to colonising bacteria or to aeroallergens. Such work should
also include functional studies exploring the potential to reverse the TH2-induced reprogramming of
virus-specific CD8+ T cells which may contribute to their pathogenic effects (Coyle, Erard et al. 1995;
Chatila, Li et al. 2008). Already the potential for CD8+ T cell-mediated immunotherapy has been
demonstrated in animal models in which antigen conjugated to cationic liposome-DNA suppressed
AHR, eosinophilia and goblet cell metaplasia through the induction of allergen-specific Tc1 cells
(Takeda, Dow et al. 2009), giving hope that human studies may be an imminent prospect.
The need for the application of deep sequencing to the study of asthma
The development of high throughput sequencing technologies (Margulies, Egholm et al. 2005) has
enabled a step-change in our ability to characterise complex microbial communities. This method was
rapidly translated from its original applications in the analysis of marine ecology to the
characterisation the human oral (Zaura, Keijser et al. 2009) and gastrointestinal (Willing, Dicksved et
al. 2010) microbiome, generating new insights into the mechanisms of complex diseases such as
inflammatory bowel disease (Willing, Dicksved et al. 2010), diabetes (Serino, Luche et al. 2012) and
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
227
obesity (Henao-Mejia, Elinav et al. 2012). Yet the respiratory community have been very slow to
adopt these tools. Indeed this year the Human Microbiome Project Consortium published data from
4788 specimens from 242 phenotyped adults, providing a reference atlas of human ecology covering
18 anatomical niches including oral, skin, lower GI and urogenital tracts but no respiratory samples
were included(Peterson, Garges et al. 2009; Nelson, Weinstock et al. 2010; 2012). To date only one
study has been published in asthma and this used the older technology of 16sRNA sequencing (Hilty,
Burke et al. 2010) rather than the whole genome sequencing approach I have used. Given the
emerging evidence implicating both the acquisition of commensal flora in early life (Bisgaard,
Hermansen et al. 2007; Thavagnanam, Fleming et al. 2008; Roduit, Scholtens et al. 2009) and the
composition of the faecal microbiome (Maeda, Noda et al. 2001; Noverr, Noggle et al. 2004;
Maslowski, Vieira et al. 2009; Bisgaard, Li et al. 2011; Ichinohe, Pang et al. 2011; Vael,
Vanheirstraeten et al. 2011; Han, Huang et al. 2012) in the pathogenesis of asthma there is an urgent
need for respiratory researcher to catch up with the rate of progress being made in other fields.
My data demonstrate the power of whole genome metagenomics to characterise the airway viral and
microbial flora in their entirety. The data argue against a hypothesised role for chronic viral
persistence in asthma (Wos, Sanak et al. 2008) and against the proposed existence of a core airway
commensal microbial community in health or in asthma (Hilty, Burke et al. 2010). Instead they
suggest that a minority of individuals with severe asthma may be suffering from chronic infection with
specific respiratory pathogens or opportunistic infections. As a consequence future research should
aim to apply this technique on a larger scale to a wide spectrum of asthmatic subjects with the aim of
defining which are the common causative organisms in this chronic infections, what are the causal
risk factors for development of these infections and what might be biomarkers to identify such
individuals in clinic in routine practice. Although some of the patients that I studied had severe
asthma, it is likely that different results, i.e. a larger microbiome, may be found in sub-phenotypes of
severe asthma, e.g. patients with chronic expectoration, patients with bronchiectasis and smoking
asthmatics who likely have elements of chronic bronchitis. Once such a comprehensive, unbiased
survey has been completed the data could then be used to produce simpler diagnostics for focussed
sets of identified pathogens, such as multiplexed PCR kits. These should then be validated in
prospective clinical trials which include antibiotics as interventions and are linked to clinically
important outcomes. Such work would likely to be of significant benefit to a small subset of subjects
with severe asthma. However, in addition community prescription of antibiotics is widespread.
Subjects in the exacerbation study had received a median 2 (IQR 1-4) doses of antibiotics in the
preceding 2 years and thus an additional and related priority for future research should be the
development of an evidence base for the rational use of antibiotic prescription for the treatment of
asthma exacerbations in the community. Again this should include the use of biomarkers and be
linked to a prospective interventional trial. My results show that raw BAL fluid has a low biomass so
future bronchoscopy studies should use either brushings or consider using ultracentrifugation to
concentrate the microbial content prior to sequencing. Finally future research into the respiratory
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
228
microbiome should include carefully planned controls for environmental contamination (Charlson,
Bittinger et al. 2011) as well as obtaining paired samples of the faecal microbiome.
MAIT cells as a priority for future research
This study is the first to investigate the role of MAIT cells in the human lung. I observed a selective
deficiency of MAIT cells in asthma, which was not related to age, but was exacerbated by systemic
corticosteroids and was subject to seasonal variation, indicating their possible regulation by vitamin D.
I established MAIT clones which allowed me to observe the heterogeneity of cytokine expression
profiles and also represent proof of concept for the ability to develop MAIT clones which will constitute
a key tool for future MAIT cell research. The high degree of evolutionary conservation of the MR1
restriction molecule (Brossay, Chioda et al. 1998; Treiner and Lantz 2006) implies these poorly
understood cells perform some key immunological functions, which are yet to be defined. The recent
discovery that the MR1 binding grove can recognise microbially-derived vitamin B metabolites (Kjer-
Nielsen, Patel et al. 2012) and their association with mucosal surfaces (Treiner, Duban et al. 2003;
Ruijing, Mengjun et al. 2011) provides strong evidence that the role of these cells is related to the
interaction between the immune system and microbes at mucosal surfaces. It is therefore likely that
they will be of relevance to airway host defence in conditions such as acute pneumonia, invasive
bacterial infection, pulmonary tuberculosis and bronchiectasis, in addition to their relationship with
severe asthma.
Future work
To conclude this thesis I will discuss future research questions, beginning with projects which I have
initiated already.
Deep sequencing of the microbiome during exacerbations
Whilst the deep sequencing data presented did not identify respiratory viruses during clinically stable
disease, such analysis of samples from subjects with symptomatic viral infections will certainly yield
very different results. Through the collaboration I have established with Prof Virgin (Washington
University School of Medicine, Saint Louis, MO) I am arranging the sequencing of samples of sputum
and nasal lavage obtained from the longitudinal study of acute viral exacerbations. Samples will be
available from all centres which participated in this multi-centre trial. Amongst 134 subjects with
clinically confirmed URITs (with a Jackson cold score >14 on two consecutive days), a virus was not
detected in 37% of nasal lavage samples tested by PCR for 21 common respiratory viruses. These
samples may contain rare or previously undiscovered pathogens. The use of an unbiased whole-
genome approach using 454 pyrosequencing or the newer Illumina HiSeq platform, coupled with a
data analysis pipeline tailored to virus pathogen discovery (Zhao) will enable characterisation of these
previously undiagnosed viral illnesses.
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
229
An integrated systems biology approach to the analysis of transcriptomic data obtained from
microarray of epithelial cells and pure T cell populations
I have obtained epithelial brushings and sorted pure populations of T cells and MAIT cells from
PBMC, sputum, and BAL which have been kept in RNA lysis buffer and sent to my collaborators
Janssen Research & Development (Springhouse, Pennsylvania) for RNA extraction and
transcriptomic analysis. To date 166 T cell samples (n=42 epithelial cells, 42 PBMC, 46 BAL, 24
sputum and 12 sputum samples after ICS) of cDNA have been extracted, passed quality control
thresholds and been successfully hybridised to the Affymetrix GeneChip Arrays. I will shortly be able
to analyse the results in collaboration with Janssen Research & Development, using pathway analysis
tools and the tranSMART knowledge management platform. This large set of paired samples from
highly phenotyped individuals will provide a powerful data-set in which to explore the distinct
activation signatures of the innate (epithelial cell), adaptive (CD3+ T cell) and innate-like (MAIT cell)
immune systems in asthma. Furthermore the pairing of the samples will enable me to explore the
interactions between these different cell types, for example the relationship between T cell cytokines
and their down-steam induction of effector pathways, and these different tissue compartments. I wish
to maximise the potential for using a fully integrated systems biology approach to generate new
hypotheses from these data-sets. Hence, through the use of the tranSMART knowledge platform I
would aim further to explore relationships between these transcriptomic data and the associated
immunological data (such as multiplex ELISA data I have already obtained from paired serum, BAL
and sputum samples) to correlate analyses at the transcriptomic and protein-levels. A further aim
would be to attempt to identify asthma endotypes through an unbiased statistical analysis of these
transcriptomic data. Together these aims constitute an ambitious undertaking, but I believe are the
essential next step in deepening our understanding of complex diseases.
A characterisation of the function of MAIT cells in human lung diseases
The data I have presented on MAIT cells provoke several questions which I wish to address in the
future.
1. Do serum levels of vitamin D3 influence MAIT cell frequencies in peripheral blood? The
seasonal variation in MAIT cell frequencies and inferences from iNKT biology (Yu and
Cantorna 2011; Yu, Zhao et al. 2011) suggests a possible relationship between serum vitamin
D3 levels and MAIT cell frequencies. I am currently awaiting the analysis by mass
spectrometry of vitamin D3 levels in 86 serum samples paired to the PBMC MAIT cell
frequencies, which may provide definitive evidence to test this hypothesis. If the data are
suggestive then it would be interesting to explore the effect of vitamin D3 on MAIT cell lines in
vitro.
2. I have found that MAIT cells are deficient in asthma. Is this due to MAIT cells migrating into the
lung during inflammation and then undergoing activation-induced apoptosis? Can these cells
recover during periods of clinical stability? If so, as these cells are readily detectable in
peripheral blood, might they be useful as a biomarker of disease activity?
Timothy SC Hinks T cell phenotypes during natural cold-induced asthma exacerbations
230
3. Are MAIT cells involved in barrier immunity in the lung? The discovery that microbial-derived
vitamin B metabolites may act as ligands for MR1 implicates MAIT cells in antimicrobial
defense. Therefore it is pertinent to investigate associations between MAIT frequencies in
chronic inflammatory lung diseases such as chronic infection, COPD and bronchiectasis.
Clinical observations could be complemented by ex vivo modeling of MAIT cell responses to
bacteria in explanted tissue. Could MAIT-targeting approaches be used therapeutically in
such chronic diseases?
4. I have found that high dose steroids can reduce MAIT cell frequencies. What is the dose-
response relationship? Can chronic low-dose ICS have a similar effect? Do steroids affect not
just the frequency, but also the function of MAIT cells?
5. What are the other functions of MAIT cells? Are they present in lung tumours, and if so, what
role do they play? Are they deficient in some patients with idiopathic bronchiectasis? Are
MAIT cells in the upper airway mucosa associated with invasive pneumococcal disease?
I have the tools necessary to investigate these cells in greater depth including antibodies for MAIT
cells and MR1 and the ability to clone MAIT cells. With the recent description of a vitamin B
metabolite 6-formyl pterin as a ligand for MR1 (Kjer-Nielsen, Patel et al. 2012) one of my supervisors,
Prof Gadola, is already developing the protocols to refold MR1 round 6-formyl pterin and other
possible ligands, which could lead to the development of another essential tool: tetramers for MAIT
cells. Thus MAIT cells constitute an emerging research area in T cell biology with the potential to
rapidly expand and we are well placed to be at the forefront of this stimulating field of research.
Timothy SC Hinks References
231
References The fat boy stood by him, breathing hard.
“My auntie told me not to run,” he explained, “on account of my asthma.”
“Ass-mar?”
“That’s right. Can’t catch me breath. I was the only boy in our school what had
asthma,” said the fat boy with a touch of pride.9
9 The protagonist - Piggy’s – childhood asthma, which was considered a rare condition at
the time, famously distanced him from his peers in Lord of the Flies, for which William
Golding won the 1983 Nobel Prize for literature. Golding, W. (1954). Lord of the Flies, Faber.
p9
Timothy SC Hinks References
232
(1991). "Workshop summary and guidelines: investigative use of bronchoscopy, lavage, and bronchial biopsies in asthma and other airway diseases." J Allergy Clin Immunol 88(5): 808-814.
(2000). "Proceedings of the ATS workshop on refractory asthma: current understanding, recommendations, and unanswered questions. American Thoracic Society." Am J Respir Crit Care Med 162(6): 2341-2351.
(2001). "British Thoracic Society guidelines on diagnostic flexible bronchoscopy." Thorax 56 Suppl 1: i1-21.
(2008). "British Guideline on the Management of Asthma." Thorax 63 Suppl 4: iv1-121. (2012). "Structure, function and diversity of the healthy human microbiome." Nature 486(7402): 207-
214. (ECACC), H. P. A. C. C. "Cell culture protocols: Procedure for Freezing Cells." Retrieved
30/10/2012, 2012, from http://www.hpacultures.org.uk/technical/ccp/procedureforfreezingcells.jsp.
(GINA), G. I. f. A. (2010). Global Strategy for Asthma Management and Prevention. Aalbers, R., H. F. Kauffman, et al. (1993). "Allergen-induced recruitment of inflammatory cells in
lavage 3 and 24 h after challenge in allergic asthmatic lungs." Chest 103(4): 1178-1184. Aalbers, R., H. F. Kauffman, et al. (1993). "Bronchial lavage and bronchoalveolar lavage in allergen-
induced single early and dual asthmatic responders." Am Rev Respir Dis 147(1): 76-81. Aberg, N., B. Hesselmar, et al. (1995). "Increase of asthma, allergic rhinitis and eczema in Swedish
schoolchildren between 1979 and 1991." Clin Exp Allergy 25(9): 815-819. Abos, B., M. Gomez Del Moral, et al. (2011). "Human MR1 expression on the cell surface is acid
sensitive, proteasome independent and increases after culturing at 26 degrees C." Biochem Biophys Res Commun 411(3): 632-636.
Acosta-Rodriguez, E. V., L. Rivino, et al. (2007). "Surface phenotype and antigenic specificity of human interleukin 17-producing T helper memory cells." Nat Immunol 8(6): 639-646.
Adamko, D. J., A. D. Fryer, et al. (2003). "CD8+ T lymphocytes in viral hyperreactivity and M2 muscarinic receptor dysfunction." Am J Respir Crit Care Med 167(4): 550-556.
Adcock, I. M., G. Caramori, et al. (2008). "New targets for drug development in asthma." Lancet 372(9643): 1073-1087.
Adura, P. (2013). Virus-host interactions following experimental rhinovirus infection in airways disease. MPhil MPhil, University of Southampton.
AgilentTechnologies (2008). Absolutely RNA FFPE Kit Instruction Manual. La Jolla, CA, Aglient Technologies.
Al-Ramli, W., D. Prefontaine, et al. (2009). "T(H)17-associated cytokines (IL-17A and IL-17F) in severe asthma." J Allergy Clin Immunol 123(5): 1185-1187.
Alcorn, J. F., C. R. Crowe, et al. (2010). "TH17 cells in asthma and COPD." Annu Rev Physiol 72: 495-516.
Amitani, R., R. Wilson, et al. (1991). "Effects of human neutrophil elastase and Pseudomonas aeruginosa proteinases on human respiratory epithelium." Am J Respir Cell Mol Biol 4(1): 26-32.
Anderson, G. P. (2008). "Endotyping asthma: new insights into key pathogenic mechanisms in a complex, heterogeneous disease." Lancet 372(9643): 1107-1119.
Anderson, G. P. and A. J. Coyle (1994). "TH2 and 'TH2-like' cells in allergy and asthma: pharmacological perspectives." Trends Pharmacol Sci 15(9): 324-332.
Angrill, J., C. Agusti, et al. (2001). "Bronchial inflammation and colonization in patients with clinically stable bronchiectasis." Am J Respir Crit Care Med 164(9): 1628-1632.
Arase, H., N. Arase, et al. (1993). "NK1.1+ CD4+ CD8- thymocytes with specific lymphokine secretion." Eur J Immunol 23(1): 307-310.
Arnoux, B., J. Bousquet, et al. (2001). "Increased bronchoalveolar lavage CD8 lymphocyte subset population in wheezy infants." Pediatr Allergy Immunol 12(4): 194-200.
Arnoux, B., D. Duval, et al. (1980). "Release of platelet-activating factor (PAF-acether) from alveolar macrophages by the calcium ionophore A23187 and phagocytosis." Eur J Clin Invest 10(6): 437-441.
Aujla, S. J., P. J. Dubin, et al. (2007). "Interleukin-17 in pulmonary host defense." Exp Lung Res 33(10): 507-518.
Timothy SC Hinks References
233
Ayyoub, M., F. Deknuydt, et al. (2009). "Human memory FOXP3+ Tregs secrete IL-17 ex vivo and constitutively express the T(H)17 lineage-specific transcription factor RORgamma t." Proc Natl Acad Sci U S A 106(21): 8635-8640.
Azzawi, M., B. Bradley, et al. (1990). "Identification of activated T lymphocytes and eosinophils in bronchial biopsies in stable atopic asthma." Am Rev Respir Dis 142(6 Pt 1): 1407-1413.
Bachert, C., P. Gevaert, et al. (2003). "IgE to Staphylococcus aureus enterotoxins in serum is related to severity of asthma." J Allergy Clin Immunol 111(5): 1131-1132.
Bachert, C., P. Gevaert, et al. (2007). "Role of staphylococcal superantigens in airway disease." Chem Immunol Allergy 93: 214-236.
Baecher-Allan, C., J. A. Brown, et al. (2001). "CD4+CD25high regulatory cells in human peripheral blood." J Immunol 167(3): 1245-1253.
Barczyk, A., W. Pierzchala, et al. (2003). "Interleukin-17 in sputum correlates with airway hyperresponsiveness to methacholine." Respir Med 97(6): 726-733.
Barnes, P. J., R. Djukanovic, et al. (2003). Respiratory Medicine. G. J. Gibson, D. M. Geddes, U. Costabel, P. J. Sterk and B. Corrin. London, Saunders. 2: 1212-1264.
Bartlett, N. W., R. P. Walton, et al. (2008). "Mouse models of rhinovirus-induced disease and exacerbation of allergic airway inflammation." Nat Med 14(2): 199-204.
Bascom, R., M. Wachs, et al. (1988). "Basophil influx occurs after nasal antigen challenge: effects of topical corticosteroid pretreatment." J Allergy Clin Immunol 81(3): 580-589.
Bateman, E. D., J. Bousquet, et al. (2008). "Stability of asthma control with regular treatment: an analysis of the Gaining Optimal Asthma controL (GOAL) study." Allergy 63(7): 932-938.
Baumgarth, N. and M. Roederer (2000). "A practical approach to multicolor flow cytometry for immunophenotyping." J Immunol Methods 243(1-2): 77-97.
Bazzi, M. D., M. A. Sultan, et al. (2011). "Interleukin 17A and F and asthma in Saudi Arabia: gene polymorphisms and protein levels." J Investig Allergol Clin Immunol 21(7): 551-555.
Beasley, R., T. Clayton, et al. (2008). "Association between paracetamol use in infancy and childhood, and risk of asthma, rhinoconjunctivitis, and eczema in children aged 6-7 years: analysis from Phase Three of the ISAAC programme." Lancet 372(9643): 1039-1048.
Belda, J., R. Leigh, et al. (2000). "Induced sputum cell counts in healthy adults." Am J Respir Crit Care Med 161(2 Pt 1): 475-478.
Belkaid, Y., C. A. Piccirillo, et al. (2002). "CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity." Nature 420(6915): 502-507.
Belkaid, Y. and B. T. Rouse (2005). "Natural regulatory T cells in infectious disease." Nat Immunol 6(4): 353-360.
Belshaw, R., V. Pereira, et al. (2004). "Long-term reinfection of the human genome by endogenous retroviruses." Proc Natl Acad Sci U S A 101(14): 4894-4899.
Bentley, A. M., Q. Meng, et al. (1993). "Increases in activated T lymphocytes, eosinophils, and cytokine mRNA expression for interleukin-5 and granulocyte/macrophage colony-stimulating factor in bronchial biopsies after allergen inhalation challenge in atopic asthmatics." Am J Respir Cell Mol Biol 8(1): 35-42.
Bentley, A. M., G. Menz, et al. (1992). "Identification of T lymphocytes, macrophages, and activated eosinophils in the bronchial mucosa in intrinsic asthma. Relationship to symptoms and bronchial responsiveness." Am Rev Respir Dis 146(2): 500-506.
Bernard, A., S. Carbonnelle, et al. (2003). "Lung hyperpermeability and asthma prevalence in schoolchildren: unexpected associations with the attendance at indoor chlorinated swimming pools." Occup Environ Med 60(6): 385-394.
Bernard, A., V. Gay-Bellile, et al. (1984). "A novel human leukocyte differentiation antigen: monoclonal antibody anti-D44 defines a 28 Kd molecule present on immature hematologic cells and a subpopulation of mature T cells." J Immunol 132(5): 2338-2344.
Betensky, R. A., E. Connick, et al. (2000). "Shipment impairs lymphocyte proliferative responses to microbial antigens." Clin Diagn Lab Immunol 7(5): 759-763.
Bettelli, E., Y. Carrier, et al. (2006). "Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells." Nature 441(7090): 235-238.
Bettelli, E., T. Korn, et al. (2008). "Induction and effector functions of T(H)17 cells." Nature 453(7198): 1051-1057.
Betts, R. J. and D. M. Kemeny (2009). "CD8+ T cells in asthma: friend or foe?" Pharmacol Ther 121(2): 123-131.
Bi, X., Y. Suzuki, et al. (2009). "High frequency and proliferation of CD4+ FOXP3+ Treg in HIV-1-infected patients with low CD4 counts." Eur J Immunol 39(1): 301-309.
Timothy SC Hinks References
234
Biagi, F., C. Badulli, et al. (2012). "Cytokine genetic profile in Whipple's disease." Eur J Clin Microbiol Infect Dis 31(11): 3145-3150.
Billerbeck, E., Y. H. Kang, et al. "Analysis of CD161 expression on human CD8+ T cells defines a distinct functional subset with tissue-homing properties." Proc Natl Acad Sci U S A 107(7): 3006-3011.
Biscione, G. L., J. Corne, et al. (2004). "Increased frequency of detection of Chlamydophila pneumoniae in asthma." Eur Respir J 24(5): 745-749.
Bisgaard, H., M. N. Hermansen, et al. (2007). "Childhood asthma after bacterial colonization of the airway in neonates." N Engl J Med 357(15): 1487-1495.
Bisgaard, H., N. Li, et al. (2011). "Reduced diversity of the intestinal microbiota during infancy is associated with increased risk of allergic disease at school age." J Allergy Clin Immunol 128(3): 646-652 e641-645.
Bizhang, M., B. Ellerbrock, et al. (2011). "Detection of nine microorganisms from the initial carious root lesions using a TaqMan-based real-time PCR." Oral Dis 17(7): 642-652.
Bjornsdottir, U. S., S. T. Holgate, et al. (2011). "Pathways activated during human asthma exacerbation as revealed by gene expression patterns in blood." PLoS ONE 6(7): e21902.
Bjornvold, M., M. C. Munthe-Kaas, et al. (2009). "A TLR2 polymorphism is associated with type 1 diabetes and allergic asthma." Genes Immun 10(2): 181-187.
Black, P. N., R. Scicchitano, et al. (2000). "Serological evidence of infection with Chlamydia pneumoniae is related to the severity of asthma." Eur Respir J 15(2): 254-259.
Blaxter, M., J. Mann, et al. (2005). "Defining operational taxonomic units using DNA barcode data." Philos Trans R Soc Lond B Biol Sci 360(1462): 1935-1943.
Bluestone, J. A. and A. K. Abbas (2003). "Natural versus adaptive regulatory T cells." Nat Rev Immunol 3(3): 253-257.
Bonelli, M., K. von Dalwigk, et al. (2008). "Foxp3 expression in CD4+ T cells of patients with systemic lupus erythematosus: a comparative phenotypic analysis." Ann Rheum Dis 67(5): 664-671.
Boniface, S., V. Koscher, et al. (2003). "Assessment of T lymphocyte cytokine production in induced sputum from asthmatics: a flow cytometry study." Clin Exp Allergy 33(9): 1238-1243.
Bousbia, S., L. Papazian, et al. (2010). "Tropheryma whipplei in patients with pneumonia." Emerg Infect Dis 16(2): 258-263.
Bousquet, J., P. K. Jeffery, et al. (2000). "Asthma. From bronchoconstriction to airways inflammation and remodeling." Am J Respir Crit Care Med 161(5): 1720-1745.
Bradding, P., J. A. Roberts, et al. (1994). "Interleukin-4, -5, and -6 and tumor necrosis factor-alpha in normal and asthmatic airways: evidence for the human mast cell as a source of these cytokines." Am J Respir Cell Mol Biol 10(5): 471-480.
Braun, R. K., C. Ferrick, et al. (2008). "IL-17 producing gammadelta T cells are required for a controlled inflammatory response after bleomycin-induced lung injury." Inflammation 31(3): 167-179.
Brightling, C. E. (2011). "Eosinophils, bronchitis and asthma: pathogenesis of cough and airflow obstruction." Pulm Pharmacol Ther 24(3): 324-327.
Brittan, J. L., T. J. Buckeridge, et al. (2012). "Pneumococcal neuraminidase A: an essential upper airway colonization factor for Streptococcus pneumoniae." Mol Oral Microbiol 27(4): 270-283.
Brossay, L., M. Chioda, et al. (1998). "CD1d-mediated recognition of an alpha-galactosylceramide by natural killer T cells is highly conserved through mammalian evolution." J Exp Med 188(8): 1521-1528.
Brusko, T., C. Wasserfall, et al. (2007). "No alterations in the frequency of FOXP3+ regulatory T-cells in type 1 diabetes." Diabetes 56(3): 604-612.
Bueno, C., J. Almeida, et al. (2001). "Flow cytometric analysis of cytokine production by normal human peripheral blood dendritic cells and monocytes: comparative analysis of different stimuli, secretion-blocking agents and incubation periods." Cytometry 46(1): 33-40.
Bullens, D. M., E. Truyen, et al. (2006). "IL-17 mRNA in sputum of asthmatic patients: linking T cell driven inflammation and granulocytic influx?" Respir Res 7: 135.
Calhoun, W. J., E. C. Dick, et al. (1994). "A common cold virus, rhinovirus 16, potentiates airway inflammation after segmental antigen bronchoprovocation in allergic subjects." J Clin Invest 94(6): 2200-2208.
Calverley, P. M., J. A. Anderson, et al. (2007). "Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease." N Engl J Med 356(8): 775-789.
Campbell, D. J. and M. A. Koch (2011). "Phenotypical and functional specialization of FOXP3+ regulatory T cells." Nat Rev Immunol 11(2): 119-130.
Timothy SC Hinks References
235
Camporota, L., A. Corkhill, et al. (2003). "The effects of Mycobacterium vaccae on allergen-induced airway responses in atopic asthma." Eur Respir J 21(2): 287-293.
Capron, M., T. Jouault, et al. (1986). "Functional study of a monoclonal antibody to IgE Fc receptor (Fc epsilon R2) of eosinophils, platelets, and macrophages." J Exp Med 164(1): 72-89.
Caraher, E. M., M. Parenteau, et al. (2000). "Flow cytometric analysis of intracellular IFN-gamma, IL-4 and IL-10 in CD3(+)4(+) T-cells from rat spleen." J Immunol Methods 244(1-2): 29-40.
Cautivo, K. M., S. M. Bueno, et al. (2010). "Efficient lung recruitment of respiratory syncytial virus-specific Th1 cells induced by recombinant bacillus Calmette-Guerin promotes virus clearance and protects from infection." J Immunol 185(12): 7633-7645.
Cembrzynska-Nowak, M., E. Szklarz, et al. (1993). "Elevated release of tumor necrosis factor-alpha and interferon-gamma by bronchoalveolar leukocytes from patients with bronchial asthma." Am Rev Respir Dis 147(2): 291-295.
Chakir, J., J. Shannon, et al. (2003). "Airway remodeling-associated mediators in moderate to severe asthma: effect of steroids on TGF-beta, IL-11, IL-17, and type I and type III collagen expression." J Allergy Clin Immunol 111(6): 1293-1298.
Chambers, E. S., A. M. Nanzer, et al. (2012). "Serum 25-dihydroxyvitamin D levels correlate with CD4(+)Foxp3(+) T-cell numbers in moderate/severe asthma." J Allergy Clin Immunol 130(2): 542-544.
Charlson, E. S., K. Bittinger, et al. (2011). "Topographical continuity of bacterial populations in the healthy human respiratory tract." Am J Respir Crit Care Med 184(8): 957-963.
Chatila, T. A., N. Li, et al. (2008). "T-cell effector pathways in allergic diseases: transcriptional mechanisms and therapeutic targets." J Allergy Clin Immunol 121(4): 812-823; quiz 824-815.
Chen, J., Y. Deng, et al. (2010). "The polymorphism of IL-17 G-152A was associated with childhood asthma and bacterial colonization of the hypopharynx in bronchiolitis." J Clin Immunol 30(4): 539-545.
Chen, M., G. Chen, et al. (2009). "Regulatory effects of IFN-beta on production of osteopontin and IL-17 by CD4+ T Cells in MS." Eur J Immunol 39(9): 2525-2536.
Chen, X., J. J. Oppenheim, et al. (2006). "Glucocorticoid amplifies IL-2-dependent expansion of functional FoxP3(+)CD4(+)CD25(+) T regulatory cells in vivo and enhances their capacity to suppress EAE." Eur J Immunol 36(8): 2139-2149.
Chen, Y., P. Thai, et al. (2003). "Stimulation of airway mucin gene expression by interleukin (IL)-17 through IL-6 paracrine/autocrine loop." J Biol Chem 278(19): 17036-17043.
Chiba, A., R. Tajima, et al. (2012). "Mucosal-associated invariant T cells promote inflammation and exacerbate disease in murine models of arthritis." Arthritis Rheum 64(1): 153-161.
Chinn, S. (2003). "Obesity and asthma: evidence for and against a causal relation." J Asthma 40(1): 1-16.
Cho, S. H., L. A. Stanciu, et al. (2005). "Increased interleukin-4, interleukin-5, and interferon-gamma in airway CD4+ and CD8+ T cells in atopic asthma." Am J Respir Crit Care Med 171(3): 224-230.
Chua, W. J., S. Kim, et al. (2011). "Endogenous MHC-related protein 1 is transiently expressed on the plasma membrane in a conformation that activates mucosal-associated invariant T cells." J Immunol 186(8): 4744-4750.
Chua, W. J., S. M. Truscott, et al. (2012). "Polyclonal mucosa-associated invariant T cells have unique innate functions in bacterial infection." Infect Immun 80(9): 3256-3267.
Coca, A. F. and R. A. Cooke (1923). "On the Classification of the Phenomena of Hypersensitiviness." J.Immunol 8: 163.
Contoli, M., S. D. Message, et al. (2006). "Role of deficient type III interferon-lambda production in asthma exacerbations." Nat Med 12(9): 1023-1026.
Cookson, W. (1999). "The alliance of genes and environment in asthma and allergy." Nature 402(6760 Suppl): B5-11.
Cookson, W. O. and M. F. Moffatt (2011). "Genetics of complex airway disease." Proc Am Thorac Soc 8(2): 149-153.
Corne, J. M., C. Marshall, et al. (2002). "Frequency, severity, and duration of rhinovirus infections in asthmatic and non-asthmatic individuals: a longitudinal cohort study." Lancet 359(9309): 831-834.
Cosmi, L., F. Annunziato, et al. (2000). "CRTH2 is the most reliable marker for the detection of circulating human type 2 Th and type 2 T cytotoxic cells in health and disease." Eur J Immunol 30(10): 2972-2979.
Cosmi, L., R. De Palma, et al. (2008). "Human interleukin 17-producing cells originate from a CD161+CD4+ T cell precursor." J Exp Med 205(8): 1903-1916.
Timothy SC Hinks References
236
Cosmi, L., L. Maggi, et al. (2010). "Identification of a novel subset of human circulating memory CD4(+) T cells that produce both IL-17A and IL-4." J Allergy Clin Immunol 125(1): 222-230 e221-224.
Costantini, A., S. Mancini, et al. (2003). "Effects of cryopreservation on lymphocyte immunophenotype and function." J Immunol Methods 278(1-2): 145-155.
Costello, E. K., C. L. Lauber, et al. (2009). "Bacterial community variation in human body habitats across space and time." Science 326(5960): 1694-1697.
Cowan, D. C., J. O. Cowan, et al. (2010). "Effects of steroid therapy on inflammatory cell subtypes in asthma." Thorax 65(5): 384-390.
Coyle, A. J., F. Erard, et al. (1995). "Virus-specific CD8+ cells can switch to interleukin 5 production and induce airway eosinophilia." J Exp Med 181(3): 1229-1233.
Crim, C., P. M. Calverley, et al. (2009). "Pneumonia risk in COPD patients receiving inhaled corticosteroids alone or in combination: TORCH study results." Eur Respir J 34(3): 641-647.
Croxford, J. L., S. Miyake, et al. (2006). "Invariant V(alpha)19i T cells regulate autoimmune inflammation." Nat Immunol 7(9): 987-994.
Cullinan, P. (2006). "Childhood allergies, birth order and family size." Thorax 61(1): 3-5. Damon, M., C. Chavis, et al. (1983). "Purification and mass spectrometry identification of leukotriene
D4 synthesized by human alveolar macrophages." Biochem Biophys Res Commun 111(2): 518-524.
Daniels, T. W., G. B. Rogers, et al. (2012). "Impact of antibiotic treatment for pulmonary exacerbations on bacterial diversity in cystic fibrosis." J Cyst Fibros.
Dardalhon, V., A. Awasthi, et al. (2008). "IL-4 inhibits TGF-beta-induced Foxp3+ T cells and, together with TGF-beta, generates IL-9+ IL-10+ Foxp3(-) effector T cells." Nat Immunol 9(12): 1347-1355.
de Pater-Huijsen, F. L., M. J. de Riemer, et al. (2002). "Products from human mast cell line cells enhance the production of interferon-gamma by CD8+ and CD4+ T cells." Immunology 106(1): 11-19.
Del Prete, G., E. Maggi, et al. (1988). "IL-4 is an essential factor for the IgE synthesis induced in vitro by human T cell clones and their supernatants." J Immunol 140(12): 4193-4198.
Dempsey, O. J. (2000). "Leukotriene receptor antagonist therapy." Postgrad Med J 76(902): 767-773. Di Lorenzo, G., M. L. Pacor, et al. (2002). "In vitro effects of fluticasone propionate on IL-13
production by mitogen-stimulated lymphocytes." Mediators Inflamm 11(3): 187-190. Diaz-Sanchez, D., L. Proietti, et al. (2003). "Diesel fumes and the rising prevalence of atopy: an urban
legend?" Curr Allergy Asthma Rep 3(2): 146-152. Disis, M. L., C. dela Rosa, et al. (2006). "Maximizing the retention of antigen specific lymphocyte
function after cryopreservation." J Immunol Methods 308(1-2): 13-18. Djukanovic, R. and S. D. Gadola (2008). "Virus infection, asthma, and chronic obstructive pulmonary
disease." N Engl J Med 359(19): 2062-2064. Djukanovic, R., P. J. Sterk, et al. (2002). "Standardised methodology of sputum induction and
processing." Eur Respir J Suppl 37: 1s-2s. Djukanovic, R., J. W. Wilson, et al. (1992). "Effect of an inhaled corticosteroid on airway inflammation
and symptoms in asthma." Am Rev Respir Dis 145(3): 669-674. Doe, C., M. Bafadhel, et al. (2010). "Expression of the T helper 17-associated cytokines IL-17A and
IL-17F in asthma and COPD." Chest 138(5): 1140-1147. Dong, Z., Z. Yang, et al. (2005). "Expression of TLR2 and TLR4 messenger RNA in the epithelial cells
of the nasal airway." Am J Rhinol 19(3): 236-239. Dougherty, R. H. and J. V. Fahy (2009). "Acute exacerbations of asthma: epidemiology, biology and
the exacerbation-prone phenotype." Clin Exp Allergy 39(2): 193-202. Droste, J. H., M. H. Wieringa, et al. (2000). "Does the use of antibiotics in early childhood increase the
risk of asthma and allergic disease?" Clin Exp Allergy 30(11): 1547-1553. Drulovic, J., E. Savic, et al. (2009). "Expression of Th1 and Th17 cytokines and transcription factors in
multiple sclerosis patients: does baseline T-bet mRNA predict the response to interferon-beta treatment?" J Neuroimmunol 215(1-2): 90-95.
Dunn, O. J. (1964). "Multiple comparisons using rank sums." Technometrics 6: 241-252. Durelli, L., L. Conti, et al. (2009). "T-helper 17 cells expand in multiple sclerosis and are inhibited by
interferon-beta." Ann Neurol 65(5): 499-509. Durham, S. R. (1998). "Mechanisms of mucosal inflammation in the nose and lungs." Clin Exp Allergy
28 Suppl 2: 11-16. Dusseaux, M., E. Martin, et al. "Human MAIT cells are xenobiotic resistant, tissue-targeted, CD161hi
IL-17 secreting T cells." Blood.
Timothy SC Hinks References
237
Dusseaux, M., E. Martin, et al. (2011). "Human MAIT cells are xenobiotic-resistant, tissue-targeted, CD161hi IL-17-secreting T cells." Blood 117(4): 1250-1259.
Eckelman, B. P., G. S. Salvesen, et al. (2006). "Human inhibitor of apoptosis proteins: why XIAP is the black sheep of the family." EMBO Rep 7(10): 988-994.
Edouard, S., F. Fenollar, et al. (2012). "The Rise of Tropheryma whipplei: a 12-Year Retrospective Study of PCR Diagnoses in Our Reference Center." J Clin Microbiol 50(12): 3917-3920.
Ege, M. J., C. Bieli, et al. (2006). "Prenatal farm exposure is related to the expression of receptors of the innate immunity and to atopic sensitization in school-age children." J Allergy Clin Immunol 117(4): 817-823.
Ege, M. J., M. Mayer, et al. (2011). "Exposure to environmental microorganisms and childhood asthma." N Engl J Med 364(8): 701-709.
Elkord, E. (2009). "Frequency of human T regulatory cells in peripheral blood is significantly reduced by cryopreservation." J Immunol Methods 347(1-2): 87-90.
Enomoto, N., E. Hyde, et al. (2012). "Allergen-specific CTL require perforin expression to suppress allergic airway inflammation." J Immunol 188(4): 1734-1741.
Erb-Downward, J. R., D. L. Thompson, et al. (2011). "Analysis of the lung microbiome in the "healthy" smoker and in COPD." PLoS ONE 6(2): e16384.
Essilfie, A.-T., J. L. Simpson, et al. (2012). "Combined Haemophilus influenzae respiratory infection and allergic airways disease drives chronic infection and features of neutrophilic asthma." Thorax.
Fenollar, F., S. Laouira, et al. (2008). "Value of Tropheryma whipplei quantitative polymerase chain reaction assay for the diagnosis of Whipple disease: usefulness of saliva and stool specimens for first-line screening." Clin Infect Dis 47(5): 659-667.
Fenollar, F., T. Ponge, et al. (2012). "First isolation of Tropheryma whipplei from bronchoalveolar fluid and clinical implications." J Infect 65(3): 275-278.
Fenollar, F., J. F. Trape, et al. (2009). "Tropheryma whipplei in fecal samples from children, Senegal." Emerg Infect Dis 15(6): 922-924.
Field, A. (2000). Discovering Statistics using SPSS for Windows. London, Sage publications. Finkelman, F. D., S. P. Hogan, et al. (2010). "Importance of cytokines in murine allergic airway
disease and human asthma." J Immunol 184(4): 1663-1674. Finotto, S., L. M. Fabbri, et al. (1991). "Increase in numbers of CD8 positive lymphocytes and
eosinophils in peripheral blood of subjects with late asthmatic reactions induced by toluene diisocyanate." Br J Ind Med 48(2): 116-121.
Fleming, D. M. and D. L. Crombie (1987). "Prevalence of asthma and hay fever in England and Wales." Br Med J (Clin Res Ed) 294(6567): 279-283.
Fleming, H. E., F. F. Little, et al. (1999). "Rhinovirus-16 colds in healthy and in asthmatic subjects: similar changes in upper and lower airways." Am J Respir Crit Care Med 160(1): 100-108.
Flint, K. C., K. B. Leung, et al. (1985). "Bronchoalveolar mast cells in extrinsic asthma: a mechanism for the initiation of antigen specific bronchoconstriction." Br Med J (Clin Res Ed) 291(6500): 923-926.
Fontenot, J. D., M. A. Gavin, et al. (2003). "Foxp3 programs the development and function of CD4+CD25+ regulatory T cells." Nat Immunol 4(4): 330-336.
Fort, M. M., J. Cheung, et al. (2001). "IL-25 induces IL-4, IL-5, and IL-13 and Th2-associated pathologies in vivo." Immunity 15(6): 985-995.
Fossiez, F., O. Djossou, et al. (1996). "T cell interleukin-17 induces stromal cells to produce proinflammatory and hematopoietic cytokines." J Exp Med 183(6): 2593-2603.
Fournier, M., F. Lebargy, et al. (1989). "Intraepithelial T-lymphocyte subsets in the airways of normal subjects and of patients with chronic bronchitis." Am Rev Respir Dis 140(3): 737-742.
Freier, E., C. S. Weber, et al. "Decrease of CD4(+)FOXP3(+) T regulatory cells in the peripheral blood of human subjects undergoing a mental stressor." Psychoneuroendocrinology 35(5): 663-673.
Frew, A. J., H. Chan, et al. (1995). "Bronchial inflammation in occupational asthma due to western red cedar." Am J Respir Crit Care Med 151(2 Pt 1): 340-344.
Frew, A. J., R. Moqbel, et al. (1990). "T lymphocytes and eosinophils in allergen-induced late-phase asthmatic reactions in the guinea pig." Am Rev Respir Dis 141(2): 407-413.
Fuhlbrigge, A., D. Peden, et al. (2012). "Asthma outcomes: exacerbations." J Allergy Clin Immunol 129(3 Suppl): S34-48.
Fujiwara, M., K. Hirose, et al. (2007). "T-bet inhibits both TH2 cell-mediated eosinophil recruitment and TH17 cell-mediated neutrophil recruitment into the airways." J Allergy Clin Immunol 119(3): 662-670.
Timothy SC Hinks References
238
Galmes, J., Y. Li, et al. (2012). "Potential implication of new torque teno mini viruses in parapneumonic empyema in children." Eur Respir J.
Ganesan, A. P. V. (2010). A study of regulatory T cells and modulation of allergic immune responses by microbial agents in human asthma. Doctor of Philosophy, University of Southampton.
Gasparich, G. E. (2010). "Spiroplasmas and phytoplasmas: microbes associated with plant hosts." Biologicals 38(2): 193-203.
Gerart, S., S. Siberil, et al. (2012). "Human iNKT and MAIT cells exhibit a PLZF-dependent pro-apoptotic propensity that is counterbalanced by XIAP." Blood.
Gern, J. E., B. Joseph, et al. (1996). "Rhinovirus inhibits antigen-specific T cell proliferation through an intercellular adhesion molecule-1-dependent mechanism." J Infect Dis 174(6): 1143-1150.
Gibson, P. G. (2007). "What do non-eosinophilic asthma and airway remodelling tell us about persistent asthma?" Thorax 62(12): 1034-1036.
Gleeson, K., D. F. Eggli, et al. (1997). "Quantitative aspiration during sleep in normal subjects." Chest 111(5): 1266-1272.
Godard, P., J. Chaintreuil, et al. (1982). "Functional assessment of alveolar macrophages: comparison of cells from asthmatics and normal subjects." J Allergy Clin Immunol 70(2): 88-93.
Godfrey, T. E., S. H. Kim, et al. (2000). "Quantitative mRNA expression analysis from formalin-fixed, paraffin-embedded tissues using 5' nuclease quantitative reverse transcription-polymerase chain reaction." J Mol Diagn 2(2): 84-91.
Gold, M. C., S. Cerri, et al. (2010). "Human mucosal associated invariant T cells detect bacterially infected cells." PLoS Biol 8(6): e1000407.
Gold, M. C., T. Eid, et al. (2012). "Human thymic MR1-restricted MAIT cells are innate pathogen-reactive effectors that adapt following thymic egress." Mucosal Immunol.
Goldstein, M. F., B. A. Veza, et al. (2001). "Comparisons of peak diurnal expiratory flow variation, postbronchodilator FEV(1) responses, and methacholine inhalation challenges in the evaluation of suspected asthma." Chest 119(4): 1001-1010.
Goleva, E., D. A. Searing, et al. (2012). "Steroid requirements and immune associations with vitamin D are stronger in children than adults with asthma." J Allergy Clin Immunol 129(5): 1243-1251.
Gould, H. J., B. J. Sutton, et al. (2003). "The biology of IGE and the basis of allergic disease." Annu Rev Immunol 21: 579-628.
Grainge, C. L., L. C. Lau, et al. (2011). "Effect of bronchoconstriction on airway remodeling in asthma." N Engl J Med 364(21): 2006-2015.
Grajewski, R. S., A. M. Hansen, et al. (2008). "Activation of invariant NKT cells ameliorates experimental ocular autoimmunity by a mechanism involving innate IFN-gamma production and dampening of the adaptive Th1 and Th17 responses." J Immunol 181(7): 4791-4797.
Green, B., V. Kehagia, et al. (2008). Pathogenic bacteria in induced sputum in severe asthma. British Thoracic Society Winter Meeting. Queen Elizabeth II Conference Centre, Westminster. 63: A49.
Green, R. H., C. E. Brightling, et al. (2002). "Asthma exacerbations and sputum eosinophil counts: a randomised controlled trial." Lancet 360(9347): 1715-1721.
Green, R. H., C. E. Brightling, et al. (2002). "Analysis of induced sputum in adults with asthma: identification of subgroup with isolated sputum neutrophilia and poor response to inhaled corticosteroids." Thorax 57(10): 875-879.
Green, R. H. and I. Pavord (2012). "Stability of inflammatory phenotypes in asthma." Thorax 67(8): 665-667.
Greenaway, H. Y., B. Ng, et al. (2012). "NKT and MAIT invariant TCRalpha sequences can be produced efficiently by VJ gene recombination." Immunobiology.
Grunberg, K., R. F. Sharon, et al. (2001). "Rhinovirus-induced airway inflammation in asthma: effect of treatment with inhaled corticosteroids before and during experimental infection." Am J Respir Crit Care Med 164(10 Pt 1): 1816-1822.
Grunberg, K., M. C. Timmers, et al. (1999). "Experimental rhinovirus 16 infection causes variable airway obstruction in subjects with atopic asthma." Am J Respir Crit Care Med 160(4): 1375-1380.
Guo, B., E. Y. Chang, et al. (2008). "The type I IFN induction pathway constrains Th17-mediated autoimmune inflammation in mice." J Clin Invest 118(5): 1680-1690.
Gupta, A., A. Bush, et al. (2012). "Vitamin d and asthma in children." Paediatr Respir Rev 13(4): 236-243.
Timothy SC Hinks References
239
Guss, A. M., G. Roeselers, et al. (2011). "Phylogenetic and metabolic diversity of bacteria associated with cystic fibrosis." ISME J 5(1): 20-29.
Gutowska-Owsiak, D., A. L. Schaupp, et al. (2012). "IL-17 downregulates filaggrin and affects keratinocyte expression of genes associated with cellular adhesion." Exp Dermatol 21(2): 104-110.
Guyot-Revol, V., J. A. Innes, et al. (2005). "Regulatory T Cells are Expanded in Blood and Disease Sites in Tuberculosis Patients." Am J Respir Crit Care Med.
Guyot-Revol, V., J. A. Innes, et al. (2006). "Regulatory T cells are expanded in blood and disease sites in patients with tuberculosis." Am J Respir Crit Care Med 173(7): 803-810.
H.A.N.C., H. A. N. C. (2011). Cross-Network PBMC Processing Standard Operating Procedure. Hachem, P., M. Lisbonne, et al. (2005). "Alpha-galactosylceramide-induced iNKT cells suppress
experimental allergic asthma in sensitized mice: role of IFN-gamma." Eur J Immunol 35(10): 2793-2802.
Hadjicharalambous, C., G. Dent, et al. (2004). "Measurement of eotaxin (CCL11) in induced sputum supernatants: validation and detection in asthma." J Allergy Clin Immunol 113(4): 657-662.
Hagman, D. K., J. N. Kuzma, et al. (2012). "Characterizing and quantifying leukocyte populations in human adipose tissue: impact of enzymatic tissue processing." J Immunol Methods.
Haldar, P. and I. D. Pavord (2007). "Noneosinophilic asthma: a distinct clinical and pathologic phenotype." J Allergy Clin Immunol 119(5): 1043-1052; quiz 1053-1044.
Haldar, P., I. D. Pavord, et al. (2008). "Cluster analysis and clinical asthma phenotypes." Am J Respir Crit Care Med 178(3): 218-224.
Hallstrand, T. S., J. D. Sprenger, et al. (2004). "Long-term acquisition of allergen-specific IgE and asthma following allogeneic bone marrow transplantation from allergic donors." Blood 104(10): 3086-3090.
Han, M. K., Y. J. Huang, et al. (2012). "Significance of the microbiome in obstructive lung disease." Thorax.
Handley, S. (2012). T. Hinks. Harju, T. H., M. Leinonen, et al. (2006). "Pathogenic bacteria and viruses in induced sputum or
pharyngeal secretions of adults with stable asthma." Thorax 61(7): 579-584. Harris, J. K., M. A. De Groote, et al. (2007). "Molecular identification of bacteria in bronchoalveolar
lavage fluid from children with cystic fibrosis." Proc Natl Acad Sci U S A 104(51): 20529-20533.
Hashimoto, K., J. E. Durbin, et al. (2005). "Respiratory syncytial virus infection in the absence of STAT 1 results in airway dysfunction, airway mucus, and augmented IL-17 levels." J Allergy Clin Immunol 116(3): 550-557.
Hashimoto, K., B. S. Graham, et al. (2004). "Respiratory syncytial virus in allergic lung inflammation increases Muc5ac and gob-5." Am J Respir Crit Care Med 170(3): 306-312.
Hawn, T. R., E. A. Misch, et al. (2007). "A common human TLR1 polymorphism regulates the innate immune response to lipopeptides." Eur J Immunol 37(8): 2280-2289.
Hawrylowicz, C., D. Richards, et al. (2002). "A defect in corticosteroid-induced IL-10 production in T lymphocytes from corticosteroid-resistant asthmatic patients." J Allergy Clin Immunol 109(2): 369-370.
Hawrylowicz, C. M. (2005). "Regulatory T cells and IL-10 in allergic inflammation." J Exp Med 202(11): 1459-1463.
Heidema, J., J. W. Rossen, et al. (2008). "Dynamics of human respiratory virus-specific CD8+ T cell responses in blood and airways during episodes of common cold." J Immunol 181(8): 5551-5559.
Heier, I., K. Malmstrom, et al. (2008). "Bronchial response pattern of antigen presenting cells and regulatory T cells in children below two years of age." Thorax.
Hellings, P. W., A. Kasran, et al. (2003). "Interleukin-17 orchestrates the granulocyte influx into airways after allergen inhalation in a mouse model of allergic asthma." Am J Respir Cell Mol Biol 28(1): 42-50.
Henao-Mejia, J., E. Elinav, et al. (2012). "Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity." Nature 482(7384): 179-185.
Herbst, T., A. Sichelstiel, et al. (2011). "Dysregulation of allergic airway inflammation in the absence of microbial colonization." Am J Respir Crit Care Med 184(2): 198-205.
Highlander, S. K. (2012). "High throughput sequencing methods for microbiome profiling: application to food animal systems." Anim Health Res Rev 13(1): 40-53.
Hill, A. T., E. J. Campbell, et al. (2000). "Association between airway bacterial load and markers of airway inflammation in patients with stable chronic bronchitis." Am J Med 109(4): 288-295.
Timothy SC Hinks References
240
Hilty, M., C. Burke, et al. (2010). "Disordered microbial communities in asthmatic airways." PLoS ONE 5(1): e8578.
Hizawa, N., M. Kawaguchi, et al. (2006). "Role of interleukin-17F in chronic inflammatory and allergic lung disease." Clin Exp Allergy 36(9): 1109-1114.
Holgate, S. T. (1999). "The epidemic of allergy and asthma." Nature 402(6760 Suppl): B2-4. Holgate, S. T. (2004). "The epidemic of asthma and allergy." J R Soc Med 97(3): 103-110. Holgate, S. T. (2008). "Pathogenesis of asthma." Clin Exp Allergy 38(6): 872-897. Holgate, S. T. (2011). "Asthma: a simple concept but in reality a complex disease." Eur J Clin Invest
41(12): 1339-1352. Holgate, S. T., P. M. Lackie, et al. (2001). "Invited lecture: activation of the epithelial mesenchymal
trophic unit in the pathogenesis of asthma." Int Arch Allergy Immunol 124(1-3): 253-258. Holgate, S. T. and R. Polosa (2006). "The mechanisms, diagnosis, and management of severe
asthma in adults." Lancet 368(9537): 780-793. Holland, P. M., R. D. Abramson, et al. (1991). "Detection of specific polymerase chain reaction
product by utilizing the 5'----3' exonuclease activity of Thermus aquaticus DNA polymerase." Proc Natl Acad Sci U S A 88(16): 7276-7280.
Holland, S. M., F. R. DeLeo, et al. (2007). "STAT3 mutations in the hyper-IgE syndrome." N Engl J Med 357(16): 1608-1619.
Holmes, A. M., R. Solari, et al. (2011). "Animal models of asthma: value, limitations and opportunities for alternative approaches." Drug Discov Today 16(15-16): 659-670.
Homer The Iliad. XV. Hori, S., T. Nomura, et al. (2003). "Control of regulatory T cell development by the transcription factor
Foxp3." Science 299(5609): 1057-1061. Howarth, P. (2012). T. Hinks. Hsieh, C. S., S. E. Macatonia, et al. (1993). "Development of TH1 CD4+ T cells through IL-12
produced by Listeria-induced macrophages." Science 260(5107): 547-549. Huang, F., C. Y. Kao, et al. (2007). "Requirement for both JAK-mediated PI3K signaling and
ACT1/TRAF6/TAK1-dependent NF-kappaB activation by IL-17A in enhancing cytokine expression in human airway epithelial cells." J Immunol 179(10): 6504-6513.
Huang, S., S. Gilfillan, et al. (2005). "Evidence for MR1 antigen presentation to mucosal-associated invariant T cells." J Biol Chem 280(22): 21183-21193.
Huang, S., S. Gilfillan, et al. (2008). "MR1 uses an endocytic pathway to activate mucosal-associated invariant T cells." J Exp Med 205(5): 1201-1211.
Huang, S., E. Martin, et al. (2009). "MR1 antigen presentation to mucosal-associated invariant T cells was highly conserved in evolution." Proc Natl Acad Sci U S A 106(20): 8290-8295.
Huang, T. J., P. A. MacAry, et al. (1999). "Effect of CD8+ T-cell depletion on bronchial hyper-responsiveness and inflammation in sensitized and allergen-exposed Brown-Norway rats." Immunology 96(3): 416-423.
Huang, Y. J., C. E. Nelson, et al. (2011). "Airway microbiota and bronchial hyperresponsiveness in patients with suboptimally controlled asthma." J Allergy Clin Immunol 127(2): 372-381 e371-373.
Huppert, J., D. Closhen, et al. (2010). "Cellular mechanisms of IL-17-induced blood-brain barrier disruption." FASEB J 24(4): 1023-1034.
Hurst, J. R., T. M. Wilkinson, et al. (2005). "Relationships among bacteria, upper airway, lower airway, and systemic inflammation in COPD." Chest 127(4): 1219-1226.
Hurst, S. D., T. Muchamuel, et al. (2002). "New IL-17 family members promote Th1 or Th2 responses in the lung: in vivo function of the novel cytokine IL-25." J Immunol 169(1): 443-453.
Hyde, H. S. (1860). On asthma, its pathology and treatment. London, J Churchill. Hymowitz, S. G., E. H. Filvaroff, et al. (2001). "IL-17s adopt a cystine knot fold: structure and activity
of a novel cytokine, IL-17F, and implications for receptor binding." EMBO J 20(19): 5332-5341.
Ichinohe, T., I. K. Pang, et al. (2011). "Microbiota regulates immune defense against respiratory tract influenza A virus infection." Proc Natl Acad Sci U S A 108(13): 5354-5359.
Illes, Z., M. Shimamura, et al. (2004). "Accumulation of Valpha7.2-Jalpha33 invariant T cells in human autoimmune inflammatory lesions in the nervous system." Int Immunol 16(2): 223-230.
Inc, A. T. (2008). Absolutely RNA Nanoprep Kit Instruction Manual. Invitrogen (2007). LIVE/DEAD Fixable Dead Cell Stain Kits. Ishigame, H., S. Kakuta, et al. (2009). "Differential roles of interleukin-17A and -17F in host defense
against mucoepithelial bacterial infection and allergic responses." Immunity 30(1): 108-119.
Timothy SC Hinks References
241
Isogai, S., A. Athiviraham, et al. (2007). "Interferon-gamma-dependent inhibition of late allergic airway responses and eosinophilia by CD8+ gammadelta T cells." Immunology 122(2): 230-238.
Ivanov, II, L. Frutos Rde, et al. (2008). "Specific microbiota direct the differentiation of IL-17-producing T-helper cells in the mucosa of the small intestine." Cell Host Microbe 4(4): 337-349.
Ivanov, II, B. S. McKenzie, et al. (2006). "The orphan nuclear receptor RORgammat directs the differentiation program of proinflammatory IL-17+ T helper cells." Cell 126(6): 1121-1133.
Jackson, G. G., H. F. Dowling, et al. (1958). "Transmission of the common cold to volunteers under controlled conditions. I. The common cold as a clinical entity." AMA Arch Intern Med 101(2): 267-278.
Jarjour, N. N., S. P. Peters, et al. (1998). "Investigative use of bronchoscopy in asthma." Am J Respir Crit Care Med 157(3 Pt 1): 692-697.
Jartti, T., M. Paul-Anttila, et al. (2009). "Systemic T-helper and T-regulatory cell type cytokine responses in rhinovirus vs. respiratory syncytial virus induced early wheezing: an observational study." Respir Res 10: 85.
Jayaram, L., M. M. Pizzichini, et al. (2006). "Determining asthma treatment by monitoring sputum cell counts: effect on exacerbations." Eur Respir J 27(3): 483-494.
Jayasekera, N. (2013). Bronchial tissue interleukin-17A and interleukin-17F expression in asthma, University of Southampton.
Jeffery, L. E., A. M. Wood, et al. (2012). "Availability of 25-Hydroxyvitamin D3 to APCs Controls the Balance between Regulatory and Inflammatory T Cell Responses." J Immunol.
Jeurink, P. V., Y. M. Vissers, et al. (2008). "T cell responses in fresh and cryopreserved peripheral blood mononuclear cells: kinetics of cell viability, cellular subsets, proliferation, and cytokine production." Cryobiology 57(2): 91-103.
Jiang, X. B., Z. D. Wang, et al. (2009). "Inhibition of CD8+ T lymphocytes attenuates respiratory syncytial virus-enhanced allergic inflammation." Respiration 77(1): 76-84.
Jiang, X. D., G. Y. Li, et al. (2011). "The characterization of IL-17A expression in patients with chronic rhinosinusitis with nasal polyps." Am J Rhinol Allergy 25(5): e171-175.
Jin, N., C. L. Roark, et al. (2009). "Allergic airway hyperresponsiveness-enhancing gammadelta T cells develop in normal untreated mice and fail to produce IL-4/13, unlike Th2 and NKT cells." J Immunol 182(4): 2002-2010.
Johnston, S. L., P. K. Pattemore, et al. (1996). "The relationship between upper respiratory infections and hospital admissions for asthma: a time-trend analysis." Am J Respir Crit Care Med 154(3 Pt 1): 654-660.
Johnston, S. L., P. K. Pattemore, et al. (1995). "Community study of role of viral infections in exacerbations of asthma in 9-11 year old children." BMJ 310(6989): 1225-1229.
Jorundsson, E., J. H. Lumsden, et al. (1999). "Rapid staining techniques in cytopathology: a review and comparison of modified protocols for hematoxylin and eosin, Papanicolaou and Romanowsky stains." Vet Clin Pathol 28(3): 100-108.
Juniper, E. F., J. Bousquet, et al. (2006). "Identifying 'well-controlled' and 'not well-controlled' asthma using the Asthma Control Questionnaire." Respir Med 100(4): 616-621.
Juniper, E. F., P. M. O'Byrne, et al. (2000). "Measuring asthma control. Clinic questionnaire or daily diary?" Am J Respir Crit Care Med 162(4 Pt 1): 1330-1334.
Juniper, E. F., K. Svensson, et al. (2005). "Measurement properties and interpretation of three shortened versions of the asthma control questionnaire." Respir Med 99(5): 553-558.
Kaiser, L., J. D. Aubert, et al. (2006). "Chronic rhinoviral infection in lung transplant recipients." Am J Respir Crit Care Med 174(12): 1392-1399.
Karagiannidis, C., M. Akdis, et al. (2004). "Glucocorticoids upregulate FOXP3 expression and regulatory T cells in asthma." J Allergy Clin Immunol 114(6): 1425-1433.
Karlsson, H. and L. Nassberger (1995). "Influence of compounds affecting synthesis, modification and transport of proteins on the expression and release of interleukin-2 receptor." Immunol Cell Biol 73(1): 81-88.
Kawachi, I., J. Maldonado, et al. (2006). "MR1-restricted V alpha 19i mucosal-associated invariant T cells are innate T cells in the gut lamina propria that provide a rapid and diverse cytokine response." J Immunol 176(3): 1618-1627.
Kawaguchi, M., D. Takahashi, et al. (2006). "IL-17F sequence variant (His161Arg) is associated with protection against asthma and antagonizes wild-type IL-17F activity." J Allergy Clin Immunol 117(4): 795-801.
Kearley, J., J. E. Barker, et al. (2005). "Resolution of airway inflammation and hyperreactivity after in vivo transfer of CD4+CD25+ regulatory T cells is interleukin 10 dependent." J Exp Med 202(11): 1539-1547.
Timothy SC Hinks References
242
Kearley, J., D. S. Robinson, et al. (2008). "CD4(+)CD25(+) regulatory T cells reverse established allergic airway inflammation and prevent airway remodeling." J Allergy Clin Immunol.
Kebir, H., K. Kreymborg, et al. (2007). "Human TH17 lymphocytes promote blood-brain barrier disruption and central nervous system inflammation." Nat Med 13(10): 1173-1175.
Keeney, E. L. (1964). "The History of Asthma from Hippocrates to Meltzer." J Allergy 35: 215-226. Keeney, M., J. W. Gratama, et al. (1998). "Isotype controls in the analysis of lymphocytes and CD34+
stem and progenitor cells by flow cytometry--time to let go!" Cytometry 34(6): 280-283. Kelly, J. T. and W. W. Busse (2008). "Host immune responses to rhinovirus: mechanisms in asthma."
J Allergy Clin Immunol 122(4): 671-682; quiz 683-674. Kepley, C. L., P. J. McFeeley, et al. (2001). "Immunohistochemical detection of human basophils in
postmortem cases of fatal asthma." Am J Respir Crit Care Med 164(6): 1053-1058. Khader, S. A., G. K. Bell, et al. (2007). "IL-23 and IL-17 in the establishment of protective pulmonary
CD4+ T cell responses after vaccination and during Mycobacterium tuberculosis challenge." Nat Immunol 8(4): 369-377.
Khader, S. A. and A. M. Cooper (2008). "IL-23 and IL-17 in tuberculosis." Cytokine 41(2): 79-83. Khattri, R., T. Cox, et al. (2003). "An essential role for Scurfin in CD4+CD25+ T regulatory cells." Nat
Immunol 4(4): 337-342. Kim, E. Y., J. T. Battaile, et al. (2008). "Persistent activation of an innate immune response translates
respiratory viral infection into chronic lung disease." Nat Med 14(6): 633-640. Kips, J. C., M. D. Inman, et al. (2002). "The use of induced sputum in clinical trials." Eur Respir J
Suppl 37: 47s-50s. Kjer-Nielsen, L., O. Patel, et al. (2012). "MR1 presents microbial vitamin B metabolites to MAIT cells."
Nature. Klemets, P., O. Lyytikainen, et al. (2010). "Risk of invasive pneumococcal infections among working
age adults with asthma." Thorax 65(8): 698-702. Kline, P. (1999). The handbook of psychological testing. London, Routledge. Kling, S., H. Donninger, et al. (2005). "Persistence of rhinovirus RNA after asthma exacerbation in
children." Clin Exp Allergy 35(5): 672-678. Kolls, J. K. and A. Linden (2004). "Interleukin-17 family members and inflammation." Immunity 21(4):
467-476. Kon, O. M., B. S. Sihra, et al. (1998). "Randomised, dose-ranging, placebo-controlled study of
chimeric antibody to CD4 (keliximab) in chronic severe asthma." Lancet 352(9134): 1109-1113.
Koshino, T., S. Teshima, et al. (1993). "Identification of basophils by immunohistochemistry in the airways of post-mortem cases of fatal asthma." Clin Exp Allergy 23(11): 919-925.
Krogh, M. (1915). "The diffusion of gases through the lungs of man." J Physiol 49(4): 271-300. Kronenberg, M. and Y. Kinjo (2009). "Innate-like recognition of microbes by invariant natural killer T
cells." Curr Opin Immunol 21(4): 391-396. Krug, N., V. J. Erpenbeck, et al. (2001). "Cytokine profile of bronchoalveolar lavage-derived CD4(+),
CD8(+), and gammadelta T cells in people with asthma after segmental allergen challenge." Am J Respir Cell Mol Biol 25(1): 125-131.
Krug, N., J. Madden, et al. (1996). "T-cell cytokine profile evaluated at the single cell level in BAL and blood in allergic asthma." Am J Respir Cell Mol Biol 14(4): 319-326.
Kudva, A., E. V. Scheller, et al. (2011). "Influenza A inhibits Th17-mediated host defense against bacterial pneumonia in mice." J Immunol 186(3): 1666-1674.
Lagier, J. C., F. Fenollar, et al. (2011). "Evidence of lifetime susceptibility to Tropheryma whipplei in patients with Whipple's disease." J Antimicrob Chemother 66(5): 1188-1189.
Laidlaw, T. M. and J. A. Boyce (2012). "Cysteinyl leukotriene receptors, old and new; implications for asthma." Clin Exp Allergy 42(9): 1313-1320.
Larche, M. (2007). "Regulatory T cells in allergy and asthma." Chest 132(3): 1007-1014. Larche, M., D. S. Robinson, et al. (2003). "The role of T lymphocytes in the pathogenesis of asthma."
J Allergy Clin Immunol 111(3): 450-463; quiz 464. Laurenzi, G. A., R. T. Potter, et al. (1961). "Bacteriologic flora of the lower respiratory tract." N Engl J
Med 265: 1273-1278. Lazarus, R., B. A. Raby, et al. (2004). "TOLL-like receptor 10 genetic variation is associated with
asthma in two independent samples." Am J Respir Crit Care Med 170(6): 594-600. Le Bourhis, L., L. Guerri, et al. (2011). "Mucosal-associated invariant T cells: unconventional
development and function." Trends Immunol 32(5): 212-218. Le Bourhis, L., E. Martin, et al. (2010). "Antimicrobial activity of mucosal-associated invariant T cells."
Nat Immunol 11(8): 701-708.
Timothy SC Hinks References
243
Lebman, D. A. and R. L. Coffman (1988). "Interleukin 4 causes isotype switching to IgE in T cell-stimulated clonal B cell cultures." J Exp Med 168(3): 853-862.
Lee, J., W. H. Ho, et al. (2001). "IL-17E, a novel proinflammatory ligand for the IL-17 receptor homolog IL-17Rh1." J Biol Chem 276(2): 1660-1664.
Lee, J. H., H. H. Yu, et al. (2007). "The levels of CD4+CD25+ regulatory T cells in paediatric patients with allergic rhinitis and bronchial asthma." Clin Exp Immunol 148(1): 53-63.
Lehmann, U. and H. Kreipe (2001). "Real-time PCR analysis of DNA and RNA extracted from formalin-fixed and paraffin-embedded biopsies." Methods 25(4): 409-418.
Leppkes, M., C. Becker, et al. (2008). "RORgamma-Expressing Th17 Cells Induce Murine Chronic Intestinal Inflammation via Redundant Effects of IL-17A and IL-17F." Gastroenterology.
Letuve, S., S. Lajoie-Kadoch, et al. (2006). "IL-17E upregulates the expression of proinflammatory cytokines in lung fibroblasts." J Allergy Clin Immunol 117(3): 590-596.
Levandowski, R. A., D. W. Ou, et al. (1986). "Acute-phase decrease of T lymphocyte subsets in rhinovirus infection." J Infect Dis 153(4): 743-748.
Lewis, S., B. Butland, et al. (1996). "Study of the aetiology of wheezing illness at age 16 in two national British birth cohorts." Thorax 51(7): 670-676.
Lewkowich, I. P., N. S. Herman, et al. (2005). "CD4+CD25+ T cells protect against experimentally induced asthma and alter pulmonary dendritic cell phenotype and function." J Exp Med 202(11): 1549-1561.
Li, H., J. Chen, et al. (2000). "Cloning and characterization of IL-17B and IL-17C, two new members of the IL-17 cytokine family." Proc Natl Acad Sci U S A 97(2): 773-778.
Lin, S. C., K. H. Chen, et al. (2007). "The quantitative analysis of peripheral blood FOXP3-expressing T cells in systemic lupus erythematosus and rheumatoid arthritis patients." Eur J Clin Invest 37(12): 987-996.
Lin, Y. L., C. C. Shieh, et al. (2008). "The functional insufficiency of human CD4+CD25 high T-regulatory cells in allergic asthma is subjected to TNF-alpha modulation." Allergy 63(1): 67-74.
Linden, A. (2001). "Role of interleukin-17 and the neutrophil in asthma." Int Arch Allergy Immunol 126(3): 179-184.
Ling, E. M., T. Smith, et al. (2004). "Relation of CD4+CD25+ regulatory T-cell suppression of allergen-driven T-cell activation to atopic status and expression of allergic disease." Lancet 363(9409): 608-615.
Liu, W., A. L. Putnam, et al. (2006). "CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells." J Exp Med 203(7): 1701-1711.
Lloyd, C. M. (2012). T. Hinks. Lloyd, C. M. and E. M. Hessel (2010). "Functions of T cells in asthma: more than just T(H)2 cells." Nat
Rev Immunol 10(12): 838-848. Lluis, A., M. Schedel, et al. (2011). "Asthma-associated polymorphisms in 17q21 influence cord blood
ORMDL3 and GSDMA gene expression and IL-17 secretion." J Allergy Clin Immunol 127(6): 1587-1594 e1586.
Lochner, M., L. Peduto, et al. (2008). "In vivo equilibrium of proinflammatory IL-17+ and regulatory IL-10+ Foxp3+ RORgamma t+ T cells." J Exp Med 205(6): 1381-1393.
Loeuillet, C., F. Martinon, et al. (2006). "Mycobacterium tuberculosis subverts innate immunity to evade specific effectors." J Immunol 177(9): 6245-6255.
Loss, G., S. Bitter, et al. (2012). "Prenatal and early-life exposures alter expression of innate immunity genes: the PASTURE cohort study." J Allergy Clin Immunol 130(2): 523-530 e529.
Lyra, A., T. Rinttila, et al. (2009). "Diarrhoea-predominant irritable bowel syndrome distinguishable by 16S rRNA gene phylotype quantification." World J Gastroenterol 15(47): 5936-5945.
Ma, C. S., G. Y. Chew, et al. (2008). "Deficiency of Th17 cells in hyper IgE syndrome due to mutations in STAT3." J Exp Med 205(7): 1551-1557.
Maalmi, H., A. Berraies, et al. (2012). "The impact of vitamin D deficiency on immune T cells in asthmatic children: a case-control study." J Asthma Allergy 5: 11-19.
Macfarlane, A. J., O. M. Kon, et al. (2000). "Basophils, eosinophils, and mast cells in atopic and nonatopic asthma and in late-phase allergic reactions in the lung and skin." J Allergy Clin Immunol 105(1 Pt 1): 99-107.
Mackenzie, G. H. (1885). "Nasal Asthma: Its Causes and Treatment." Br Med J 1(1272): 984-985. Maecker, H. T., J. Moon, et al. (2005). "Impact of cryopreservation on tetramer, cytokine flow
cytometry, and ELISPOT." BMC Immunol 6: 17.
Timothy SC Hinks References
244
Maeda, Y., S. Noda, et al. (2001). "The failure of oral tolerance induction is functionally coupled to the absence of T cells in Peyer's patches under germfree conditions." Immunobiology 204(4): 442-457.
Maggi, F., M. Pifferi, et al. (2003). "TT virus in the nasal secretions of children with acute respiratory diseases: relations to viremia and disease severity." J Virol 77(4): 2418-2425.
Magnan, A. O., L. G. Mely, et al. (2000). "Assessment of the Th1/Th2 paradigm in whole blood in atopy and asthma. Increased IFN-gamma-producing CD8(+) T cells in asthma." Am J Respir Crit Care Med 161(6): 1790-1796.
Malmstrom, K., A. Pitkaranta, et al. (2006). "Human rhinovirus in bronchial epithelium of infants with recurrent respiratory symptoms." J Allergy Clin Immunol 118(3): 591-596.
Mamessier, E., A. M. Lorec, et al. (2007). "T regulatory cells in stable posttransplant bronchiolitis obliterans syndrome." Transplantation 84(7): 908-916.
Mamessier, E., F. Milhe, et al. (2007). "T-cell activation in occupational asthma and rhinitis." Allergy 62(2): 162-169.
Mamessier, E., A. Nieves, et al. (2008). "T-cell activation during exacerbations: a longitudinal study in refractory asthma." Allergy 63(9): 1202-1210.
Margulies, M., M. Egholm, et al. (2005). "Genome sequencing in microfabricated high-density picolitre reactors." Nature 437(7057): 376-380.
Marks, G. B., K. Ng, et al. (2003). "The effect of neonatal BCG vaccination on atopy and asthma at age 7 to 14 years: an historical cohort study in a community with a very low prevalence of tuberculosis infection and a high prevalence of atopic disease." J Allergy Clin Immunol 111(3): 541-549.
Martin-Saavedra, F. M., C. Gonzalez-Garcia, et al. (2008). "Beta interferon restricts the inflammatory potential of CD4+ cells through the boost of the Th2 phenotype, the inhibition of Th17 response and the prevalence of naturally occurring T regulatory cells." Mol Immunol 45(15): 4008-4019.
Martin, E., E. Treiner, et al. (2009). "Stepwise development of MAIT cells in mouse and human." PLoS Biol 7(3): e54.
Maslowski, K. M., A. T. Vieira, et al. (2009). "Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43." Nature 461(7268): 1282-1286.
Masuda, N., T. Ohnishi, et al. (1999). "Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples." Nucleic Acids Res 27(22): 4436-4443.
Matsumoto, K., H. Inoue, et al. (2004). "Decrease of interleukin-10-producing T cells in the peripheral blood of severe unstable atopic asthmatics." Int Arch Allergy Immunol 134(4): 295-302.
Matsumoto, K., H. Inoue, et al. (2008). "Different profiles of IL-10+IFN-gamma-IL-4-CD4+ T cells in the peripheral blood in atopic and non-atopic asthmatics." Respiration 75(3): 281-287.
Matthews, J. N., D. G. Altman, et al. (1990). "Analysis of serial measurements in medical research." BMJ 300(6719): 230-235.
Matulis, G., J. P. Sanderson, et al. (2010). "Innate-like control of human iNKT cell autoreactivity via the hypervariable CDR3beta loop." PLoS Biol 8(6): e1000402.
McKinley, L., J. F. Alcorn, et al. (2008). "TH17 cells mediate steroid-resistant airway inflammation and airway hyperresponsiveness in mice." J Immunol 181(6): 4089-4097.
McLoughlin, R. M., A. Calatroni, et al. (2012). "Longitudinal relationship of early life immunomodulatory T cell phenotype and function to development of allergic sensitization in an urban cohort." Clin Exp Allergy 42(3): 392-404.
McNicholl, D. M., M. Stevenson, et al. (2012). "The utility of fractional exhaled nitric oxide suppression in the identification of nonadherence in difficult asthma." Am J Respir Crit Care Med 186(11): 1102-1108.
Meltzer, S. J. (1910). "Bronchial asthma as a Phenomenon of Anaphylaxis." J.A.M.A 55: 1021. Message, S. D. and S. L. Johnston (2001). "The immunology of virus infection in asthma." Eur Respir
J 18(6): 1013-1025. Message, S. D., V. Laza-Stanca, et al. (2008). "Rhinovirus-induced lower respiratory illness is
increased in asthma and related to virus load and Th1/2 cytokine and IL-10 production." Proc Natl Acad Sci U S A 105(36): 13562-13567.
Meyer, E. H., R. H. DeKruyff, et al. (2008). "T cells and NKT cells in the pathogenesis of asthma." Annu Rev Med 59: 281-292.
Michel, M. L., D. Mendes-da-Cruz, et al. (2008). "Critical role of ROR-gammat in a new thymic pathway leading to IL-17-producing invariant NKT cell differentiation." Proc Natl Acad Sci U S A 105(50): 19845-19850.
Timothy SC Hinks References
245
Miller, M. R., J. Hankinson, et al. (2005). "Standardisation of spirometry." Eur Respir J 26(2): 319-338. Mills, K. H. (2004). "Regulatory T cells: friend or foe in immunity to infection?" Nat Rev Immunol 4(11):
841-855. Milner, J. D., J. M. Brenchley, et al. (2008). "Impaired T(H)17 cell differentiation in subjects with
autosomal dominant hyper-IgE syndrome." Nature 452(7188): 773-776. Mitchell, C., K. Provost, et al. (2011). "IFN-gamma acts on the airway epithelium to inhibit local and
systemic pathology in allergic airway disease." J Immunol 187(7): 3815-3820. Miyazaki, Y., S. Miyake, et al. (2011). "Mucosal-associated invariant T cells regulate Th1 response in
multiple sclerosis." Int Immunol 23(9): 529-535. Moffatt, M. F., M. R. Hill, et al. (1994). "Genetic linkage of T-cell receptor alpha/delta complex to
specific IgE responses." Lancet 343(8913): 1597-1600. Molet, S., Q. Hamid, et al. (2001). "IL-17 is increased in asthmatic airways and induces human
bronchial fibroblasts to produce cytokines." J Allergy Clin Immunol 108(3): 430-438. Mollenhauer, H. H., D. J. Morre, et al. (1990). "Alteration of intracellular traffic by monensin;
mechanism, specificity and relationship to toxicity." Biochim Biophys Acta 1031(2): 225-246. Moller-Larsen, S., M. Nyegaard, et al. (2008). "Association analysis identifies TLR7 and TLR8 as
novel risk genes in asthma and related disorders." Thorax 63(12): 1064-1069. Monk, P. (2012). T. Hinks. Moore, W. C., D. A. Meyers, et al. (2010). "Identification of asthma phenotypes using cluster analysis
in the Severe Asthma Research Program." Am J Respir Crit Care Med 181(4): 315-323. Morens, D. M., J. K. Taubenberger, et al. (2008). "Predominant role of bacterial pneumonia as a
cause of death in pandemic influenza: implications for pandemic influenza preparedness." J Infect Dis 198(7): 962-970.
Moseley, T. A., D. R. Haudenschild, et al. (2003). "Interleukin-17 family and IL-17 receptors." Cytokine Growth Factor Rev 14(2): 155-174.
Mosmann, T. R., H. Cherwinski, et al. (1986). "Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins." J Immunol 136(7): 2348-2357.
Mosmann, T. R., L. Li, et al. (1997). "Functions of CD8 T-cell subsets secreting different cytokine patterns." Semin Immunol 9(2): 87-92.
Mucida, D., Y. Park, et al. (2007). "Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid." Science 317(5835): 256-260.
Mumberg, D., P. A. Monach, et al. (1999). "CD4(+) T cells eliminate MHC class II-negative cancer cells in vivo by indirect effects of IFN-gamma." Proc Natl Acad Sci U S A 96(15): 8633-8638.
Murakami, M., A. Sakamoto, et al. (2002). "CD25+CD4+ T cells contribute to the control of memory CD8+ T cells." Proc Natl Acad Sci U S A 99(13): 8832-8837.
Murdoch, J. R. and C. M. Lloyd (2010). "Resolution of allergic airway inflammation and airway hyperreactivity is mediated by IL-17-producing {gamma}{delta}T cells." Am J Respir Crit Care Med 182(4): 464-476.
Murphy, K., P. Travers, et al. (2008). Janeway's Immunobiology. New York, Garland Science. Mutalithas, K., J. Croudace, et al. (2007). "Bronchoalveolar lavage invariant natural killer T cells are
not increased in asthma." J Allergy Clin Immunol 119(5): 1274-1276. Naseer, T., E. M. Minshall, et al. (1997). "Expression of IL-12 and IL-13 mRNA in asthma and their
modulation in response to steroid therapy." Am J Respir Crit Care Med 155(3): 845-851. Nelson, D. R., Lehninger, A.L., Cox, M. (2005). Lehninger principles of biochemistry. New York, W.H.
Freeman. Nelson, K. E., G. M. Weinstock, et al. (2010). "A catalog of reference genomes from the human
microbiome." Science 328(5981): 994-999. Nicholas, B., K. J. Staples, et al. (2013). Vacuolar ATPases as host targets for anti-influenza drugs,
University of Southampton. Noverr, M. C., R. M. Noggle, et al. (2004). "Role of antibiotics and fungal microbiota in driving
pulmonary allergic responses." Infect Immun 72(9): 4996-5003. Nylander, S. and I. Kalies (1999). "Brefeldin A, but not monensin, completely blocks CD69 expression
on mouse lymphocytes: efficacy of inhibitors of protein secretion in protocols for intracellular cytokine staining by flow cytometry." J Immunol Methods 224(1-2): 69-76.
O'Connor, W., Jr., M. Kamanaka, et al. (2009). "A protective function for interleukin 17A in T cell-mediated intestinal inflammation." Nat Immunol 10(6): 603-609.
O'Donnell, R., D. Breen, et al. (2006). "Inflammatory cells in the airways in COPD." Thorax 61(5): 448-454.
Timothy SC Hinks References
246
O'Neil-Andersen, N. J. and D. A. Lawrence (2002). "Differential modulation of surface and intracellular protein expression by T cells after stimulation in the presence of monensin or brefeldin A." Clin Diagn Lab Immunol 9(2): 243-250.
O'Shaughnessy, T. C., T. W. Ansari, et al. (1997). "Inflammation in bronchial biopsies of subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocytes with FEV1." Am J Respir Crit Care Med 155(3): 852-857.
O'Sullivan, S., L. Cormican, et al. (2001). "Activated, cytotoxic CD8(+) T lymphocytes contribute to the pathology of asthma death." Am J Respir Crit Care Med 164(4): 560-564.
Oda, N., P. B. Canelos, et al. (2005). "Interleukin-17F induces pulmonary neutrophilia and amplifies antigen-induced allergic response." Am J Respir Crit Care Med 171(1): 12-18.
Ooi, J. H., J. Chen, et al. (2012). "Vitamin D regulation of immune function in the gut: why do T cells have vitamin D receptors?" Mol Aspects Med 33(1): 77-82.
Orgun, N. N., M. A. Mathis, et al. (2008). "Deviation from a strong Th1-dominated to a modest Th17-dominated CD4 T cell response in the absence of IL-12p40 and type I IFNs sustains protective CD8 T cells." J Immunol 180(6): 4109-4115.
Ostroukhova, M., Z. Qi, et al. (2006). "Treg-mediated immunosuppression involves activation of the Notch-HES1 axis by membrane-bound TGF-beta." J Clin Invest 116(4): 996-1004.
Owen, R. E., E. Sinclair, et al. (2007). "Loss of T cell responses following long-term cryopreservation." J Immunol Methods 326(1-2): 93-115.
Pachlopnik Schmid, J., D. Canioni, et al. (2011). "Clinical similarities and differences of patients with X-linked lymphoproliferative syndrome type 1 (XLP-1/SAP deficiency) versus type 2 (XLP-2/XIAP deficiency)." Blood 117(5): 1522-1529.
Page, K., K. M. Lierl, et al. (2008). "TLR2-mediated activation of neutrophils in response to German cockroach frass." J Immunol 180(9): 6317-6324.
Panzani, R. C. (1988). "Seneca and his asthma: the illnesses, life, and death of a Roman stoic philosopher." J Asthma 25(3): 163-174.
Papadopouli, E., N. Tzanakis, et al. (2006). "Comparison of induced sputum inflammatory profiles between childhood and adult-onset asthma." Respir Med 100(8): 1442-1450.
Park, H., Z. Li, et al. (2005). "A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17." Nat Immunol 6(11): 1133-1141.
Park, H. Y., M. Kim, et al. (2011). "Stereospecific microbial production of isoflavanones from isoflavones and isoflavone glucosides." Appl Microbiol Biotechnol 91(4): 1173-1181.
Park, S. J. and Y. C. Lee (2010). "Interleukin-17 regulation: an attractive therapeutic approach for asthma." Respir Res 11: 78.
Parry, D. E., W. W. Busse, et al. (2000). "Rhinovirus-induced PBMC responses and outcome of experimental infection in allergic subjects." J Allergy Clin Immunol 105(4): 692-698.
Paul, G., J. M. Brehm, et al. (2012). "Vitamin D and asthma." Am J Respir Crit Care Med 185(2): 124-132.
Pavord, I. D., C. E. Brightling, et al. (1999). "Non-eosinophilic corticosteroid unresponsive asthma." Lancet 353(9171): 2213-2214.
Pecora, D. V. (1963). "A Comparison of Transtracheal Aspiration with Other Methods of Determining the Bacterial Flora of the Lower Respiratory Tract." N Engl J Med 269: 664-666.
Pene, J., S. Chevalier, et al. (2008). "Chronically inflamed human tissues are infiltrated by highly differentiated Th17 lymphocytes." J Immunol 180(11): 7423-7430.
Perttunen, H., E. Moilanen, et al. (2008). "Beta2-agonists potentiate corticosteroid-induced neutrophil survival." COPD 5(3): 163-169.
Peterfalvi, A., E. Gomori, et al. (2008). "Invariant Valpha7.2-Jalpha33 TCR is expressed in human kidney and brain tumors indicating infiltration by mucosal-associated invariant T (MAIT) cells." Int Immunol 20(12): 1517-1525.
Peterson, J., S. Garges, et al. (2009). "The NIH Human Microbiome Project." Genome Res 19(12): 2317-2323.
Pichavant, M., S. Goya, et al. (2008). "Ozone exposure in a mouse model induces airway hyperreactivity that requires the presence of natural killer T cells and IL-17." J Exp Med 205(2): 385-393.
Pierrou, S., P. Broberg, et al. (2007). "Expression of genes involved in oxidative stress responses in airway epithelial cells of smokers with chronic obstructive pulmonary disease." Am J Respir Crit Care Med 175(6): 577-586.
Porcelli, S., C. E. Yockey, et al. (1993). "Analysis of T cell antigen receptor (TCR) expression by human peripheral blood CD4-8- alpha/beta T cells demonstrates preferential use of several V beta genes and an invariant TCR alpha chain." J Exp Med 178(1): 1-16.
Timothy SC Hinks References
247
Prause, O., S. Bozinovski, et al. (2004). "Increased matrix metalloproteinase-9 concentration and activity after stimulation with interleukin-17 in mouse airways." Thorax 59(4): 313-317.
PrimerDesignLtd. "PerfectProbe; An improved Real-Time PCR Probe." Retrieved 02/11/2012, 2012, from http://www.primerdesign.co.uk/perfect_probe.html.
PrimerDesignLtd Precision nanoScript Reverse Transcription kit Instructions for cDNA synthesis, Primer Design.
Provoost, S., T. Maes, et al. (2009). "Decreased FOXP3 protein expression in patients with asthma." Allergy 64(10): 1539-1546.
Quan, S. H., Y. L. Zhang, et al. (2012). "Contribution of interleukin 17A to the development and regulation of allergic inflammation in a murine allergic rhinitis model." Ann Allergy Asthma Immunol 108(5): 342-350.
Quanjer, P. H., G. J. Tammeling, et al. (1993). "Lung volumes and forced ventilatory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society." Eur Respir J Suppl 16: 5-40.
Quintana, F. J., A. S. Basso, et al. (2008). "Control of T(reg) and T(H)17 cell differentiation by the aryl hydrocarbon receptor." Nature 453(7191): 65-71.
Rachitskaya, A. V., A. M. Hansen, et al. (2008). "Cutting edge: NKT cells constitutively express IL-23 receptor and RORgammat and rapidly produce IL-17 upon receptor ligation in an IL-6-independent fashion." J Immunol 180(8): 5167-5171.
Ramgolam, V. S., Y. Sha, et al. (2009). "IFN-beta inhibits human Th17 cell differentiation." J Immunol 183(8): 5418-5427.
Randolph, D. A., R. Stephens, et al. (1999). "Cooperation between Th1 and Th2 cells in a murine model of eosinophilic airway inflammation." J Clin Invest 104(8): 1021-1029.
Ratnaike, R. N. (2000). "Whipple's disease." Postgrad Med J 76(902): 760-766. Reantragoon, R., L. Kjer-Nielsen, et al. (2012). "Structural insight into MR1-mediated recognition of
the mucosal associated invariant T cell receptor." The Journal of Experimental Medicine. Reinherz, E. L., P. C. Kung, et al. (1980). "Discrete stages of human intrathymic differentiation:
analysis of normal thymocytes and leukemic lymphoblasts of T-cell lineage." Proc Natl Acad Sci U S A 77(3): 1588-1592.
Reinherz, E. L. and S. F. Schlossman (1980). "The differentiation and function of human T lymphocytes." Cell 19(4): 821-827.
Reynolds, J. M., P. Angkasekwinai, et al. (2010). "IL-17 family member cytokines: regulation and function in innate immunity." Cytokine Growth Factor Rev 21(6): 413-423.
Richards, D. F., M. Fernandez, et al. (2000). "Glucocorticoids drive human CD8(+) T cell differentiation towards a phenotype with high IL-10 and reduced IL-4, IL-5 and IL-13 production." Eur J Immunol 30(8): 2344-2354.
Riedler, J., C. Braun-Fahrlander, et al. (2001). "Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey." Lancet 358(9288): 1129-1133.
Rietz, H., A. L. Plummer, et al. (2002). "Asthma as a consequence of bone marrow transplantation." Chest 122(1): 369-370.
Robinson, D. S., Q. Hamid, et al. (1992). "Predominant TH2-like bronchoalveolar T-lymphocyte population in atopic asthma." N Engl J Med 326(5): 298-304.
Robinson, D. S., M. Larche, et al. (2004). "Tregs and allergic disease." J Clin Invest 114(10): 1389-1397.
Roduit, C., S. Scholtens, et al. (2009). "Asthma at 8 years of age in children born by caesarean section." Thorax 64(2): 107-113.
Roederer, M. (2001). "Spectral compensation for flow cytometry: visualization artifacts, limitations, and caveats." Cytometry 45(3): 194-205.
Rolain, J. M., F. Fenollar, et al. (2007). "False positive PCR detection of Tropheryma whipplei in the saliva of healthy people." BMC Microbiol 7: 48.
Rona, R. J., S. Chinn, et al. (1995). "Trends in the prevalence of asthma in Scottish and English primary school children 1982-92." Thorax 50(9): 992-993.
Ronaghi, M. (2001). "Pyrosequencing sheds light on DNA sequencing." Genome Res 11(1): 3-11. Roncador, G., P. J. Brown, et al. (2005). "Analysis of FOXP3 protein expression in human
CD4+CD25+ regulatory T cells at the single-cell level." Eur J Immunol 35(6): 1681-1691. Roncarolo, M. G. and M. Battaglia (2007). "Regulatory T-cell immunotherapy for tolerance to self
antigens and alloantigens in humans." Nat Rev Immunol 7(8): 585-598. Rubin, E. H. and M. Rubin (1947). Diseases of the Chest. Philadelphia and London, W.B. Saunders. Ruijing, X., W. Mengjun, et al. (2011). "Ja33+ MAIT Cells Play a Protective Role in TNBS Induced
Intestinal Inflammation." Hepatogastroenterology 59(115-116).
Timothy SC Hinks References
248
Ruijing, X., W. Mengjun, et al. (2012). "Jalpha33+ MAIT cells play a protective role in TNBS induced intestinal inflammation." Hepatogastroenterology 59(115): 762-767.
Ruitenberg, J. J., C. B. Mulder, et al. (2006). "VACUTAINER CPT and Ficoll density gradient separation perform equivalently in maintaining the quality and function of PBMC from HIV seropositive blood samples." BMC Immunol 7: 11.
Russell, S. L., M. J. Gold, et al. (2012). "Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma." EMBO Rep 13(5): 440-447.
Sabroe, I., L. C. Parker, et al. (2007). "Targeting the networks that underpin contiguous immunity in asthma and chronic obstructive pulmonary disease." Am J Respir Crit Care Med 175(4): 306-311.
Saetta, M., A. Di Stefano, et al. (1998). "CD8+ T-lymphocytes in peripheral airways of smokers with chronic obstructive pulmonary disease." Am J Respir Crit Care Med 157(3 Pt 1): 822-826.
Saffar, A. S., H. Ashdown, et al. (2011). "The molecular mechanisms of glucocorticoids-mediated neutrophil survival." Curr Drug Targets 12(4): 556-562.
Saitoh, T., T. Kusunoki, et al. (2010). "Role of interleukin-17A in the eosinophil accumulation and mucosal remodeling in chronic rhinosinusitis with nasal polyps associated with asthma." Int Arch Allergy Immunol 151(1): 8-16.
Sajjan, U., Q. Wang, et al. (2008). "Rhinovirus disrupts the barrier function of polarized airway epithelial cells." Am J Respir Crit Care Med 178(12): 1271-1281.
Sakaguchi, S., N. Sakaguchi, et al. (1995). "Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases." J Immunol 155(3): 1151-1164.
Sakula, A. (1985). "Henry Hyde Salter (1823-71): a biographical sketch." Thorax 40(12): 887-888. Sanderson, C. J. (1992). "Interleukin-5, eosinophils, and disease." Blood 79(12): 3101-3109. Sano, K., K. Haneda, et al. (1999). "Ovalbumin (OVA) and Mycobacterium tuberculosis bacilli
cooperatively polarize anti-OVA T-helper (Th) cells toward a Th1-dominant phenotype and ameliorate murine tracheal eosinophilia." Am J Respir Cell Mol Biol 20(6): 1260-1267.
Sapkota, A. R., S. Berger, et al. (2010). "Human pathogens abundant in the bacterial metagenome of cigarettes." Environ Health Perspect 118(3): 351-356.
Sawicka, E., A. Noble, et al. (2004). "Tc2 cells respond to soluble antigen in the respiratory tract and induce lung eosinophilia and bronchial hyperresponsiveness." Eur J Immunol 34(9): 2599-2608.
Schinnerling, K., V. Moos, et al. (2011). "Regulatory T cells in patients with Whipple's disease." J Immunol 187(8): 4061-4067.
Schmidt-Weber, C. B., M. Akdis, et al. (2007). "TH17 cells in the big picture of immunology." J Allergy Clin Immunol 120(2): 247-254.
Schnyder-Candrian, S., D. Togbe, et al. (2006). "Interleukin-17 is a negative regulator of established allergic asthma." J Exp Med 203(12): 2715-2725.
Schuerwegh, A. J., W. J. Stevens, et al. (2001). "Evaluation of monensin and brefeldin A for flow cytometric determination of interleukin-1 beta, interleukin-6, and tumor necrosis factor-alpha in monocytes." Cytometry 46(3): 172-176.
Schulze, J., M. Rosewich, et al. (2009). "Methacholine challenge--comparison of an ATS protocol to a new rapid single concentration technique." Respir Med 103(12): 1898-1903.
Schwandner, R., K. Yamaguchi, et al. (2000). "Requirement of tumor necrosis factor receptor-associated factor (TRAF)6 in interleukin 17 signal transduction." J Exp Med 191(7): 1233-1240.
Schwarze, J., G. Cieslewicz, et al. (1999). "CD8 T cells are essential in the development of respiratory syncytial virus-induced lung eosinophilia and airway hyperresponsiveness." J Immunol 162(7): 4207-4211.
Seddiki, N., B. Santner-Nanan, et al. (2006). "Expression of interleukin (IL)-2 and IL-7 receptors discriminates between human regulatory and activated T cells." J Exp Med 203(7): 1693-1700.
Seissler, N., E. Schmitt, et al. (2012). "Methylprednisolone treatment increases the proportion of the highly suppressive HLA-DR(+)-Treg-cells in transplanted patients." Transpl Immunol 27(4): 157-161.
Semik-Orzech, A., A. Barczyk, et al. (2009). "Interleukin 17 and RANTES levels in induced sputum of patients with allergic rhinitis after a single nasal allergen challenge." Ann Allergy Asthma Immunol 103(5): 418-424.
Seneca, L. A. (65-65 AD). Epistulae Morales ad Lucilium.
Timothy SC Hinks References
249
Sergejeva, S., S. Ivanov, et al. (2005). "Interleukin-17 as a recruitment and survival factor for airway macrophages in allergic airway inflammation." Am J Respir Cell Mol Biol 33(3): 248-253.
Serino, M., E. Luche, et al. (2012). "Metabolic adaptation to a high-fat diet is associated with a change in the gut microbiota." Gut 61(4): 543-553.
Sethi, S., N. Evans, et al. (2002). "New strains of bacteria and exacerbations of chronic obstructive pulmonary disease." N Engl J Med 347(7): 465-471.
Shevach, E. M. (2006). "From vanilla to 28 flavors: multiple varieties of T regulatory cells." Immunity 25(2): 195-201.
Shi, Y. H., G. C. Shi, et al. (2011). "Coexistence of Th1/Th2 and Th17/Treg imbalances in patients with allergic asthma." Chin Med J (Engl) 124(13): 1951-1956.
Shinohara, M. L., J. H. Kim, et al. (2008). "Engagement of the type I interferon receptor on dendritic cells inhibits T helper 17 cell development: role of intracellular osteopontin." Immunity 29(1): 68-78.
Silver, R. F., L. Zukowski, et al. (2003). "Recruitment of antigen-specific Th1-like responses to the human lung following bronchoscopic segmental challenge with purified protein derivative of Mycobacterium tuberculosis." Am J Respir Cell Mol Biol 29(1): 117-123.
Simione, M. S. (1998). Cryopreservation Manual. A. T. C. C. (ATCC), Nalge Nunc International Corp. Simpson, J. L., T. V. Grissell, et al. (2007). "Innate immune activation in neutrophilic asthma and
bronchiectasis." Thorax 62(3): 211-218. Simpson, J. L., S. Phipps, et al. (2009). "Inflammatory mechanisms and treatment of obstructive
airway diseases with neutrophilic bronchitis." Pharmacol Ther 124(1): 86-95. Simpson, J. L., H. Powell, et al. (2008). "Clarithromycin targets neutrophilic airway inflammation in
refractory asthma." Am J Respir Crit Care Med 177(2): 148-155. Simpson, J. L., R. Scott, et al. (2006). "Inflammatory subtypes in asthma: assessment and
identification using induced sputum." Respirology 11(1): 54-61. Singh, D., L. Edwards, et al. (2010). "Sputum neutrophils as a biomarker in COPD: findings from the
ECLIPSE study." Respir Res 11: 77. Song, C., L. Luo, et al. (2008). "IL-17-Producing Alveolar Macrophages Mediate Allergic Lung
Inflammation Related to Asthma." J Immunol 181(9): 6117-6124. Sorkness, R. L., G. Gonzalez-Fernandez, et al. (2008). "The asthma index: a continuous variable to
characterize exacerbations of asthma." J Allergy Clin Immunol 122(4): 838-840. Soyka, M. B., P. Wawrzyniak, et al. (2012). "Defective epithelial barrier in chronic rhinosinusitis: The
regulation of tight junctions by IFN-gamma and IL-4." J Allergy Clin Immunol 130(5): 1087-1096 e1010.
Soysal, A., N. Bahceciler, et al. (2008). "Lack of an inverse association between tuberculosis infection and atopy: by T-cell-based immune assay (RD1-ELISpot)." Pediatr Allergy Immunol 19(8): 709-715.
Spanevello, A., M. Confalonieri, et al. (2000). "Induced sputum cellularity. Reference values and distribution in normal volunteers." Am J Respir Crit Care Med 162(3 Pt 1): 1172-1174.
Sporik, R., S. T. Holgate, et al. (1990). "Exposure to house-dust mite allergen (Der p I) and the development of asthma in childhood. A prospective study." N Engl J Med 323(8): 502-507.
Staley, J. T. and A. Konopka (1985). "Measurement of in situ activities of nonphotosynthetic microorganisms in aquatic and terrestrial habitats." Annu Rev Microbiol 39: 321-346.
Staples, K. J., T. S. Hinks, et al. (2012). "Phenotypic characterization of lung macrophages in asthmatic patients: Overexpression of CCL17." J Allergy Clin Immunol.
Stark, M. A., Y. Huo, et al. (2005). "Phagocytosis of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL-17." Immunity 22(3): 285-294.
Starnes, T., M. J. Robertson, et al. (2001). "Cutting edge: IL-17F, a novel cytokine selectively expressed in activated T cells and monocytes, regulates angiogenesis and endothelial cell cytokine production." J Immunol 167(8): 4137-4140.
Strachan, D. P. (1989). "Hay fever, hygiene, and household size." BMJ 299(6710): 1259-1260. Strachan, D. P. (1997). "Allergy and family size: a riddle worth solving." Clin Exp Allergy 27(3): 235-
236. Strachan, D. P. (2000). "Family size, infection and atopy: the first decade of the "hygiene
hypothesis"." Thorax 55 Suppl 1: S2-10. Strauss, E. (2009). "Microbiology. Phytoplasma research begins to bloom." Science 325(5939): 388-
390. Strickland, D. H., P. A. Stumbles, et al. (2006). "Reversal of airway hyperresponsiveness by induction
of airway mucosal CD4+CD25+ regulatory T cells." J Exp Med 203(12): 2649-2660.
Timothy SC Hinks References
250
Sur, S., T. B. Crotty, et al. (1993). "Sudden-onset fatal asthma. A distinct entity with few eosinophils and relatively more neutrophils in the airway submucosa?" Am Rev Respir Dis 148(3): 713-719.
Suzuki, S., F. Kokubu, et al. (2007). "Expression of interleukin-17F in a mouse model of allergic asthma." Int Arch Allergy Immunol 143 Suppl 1: 89-94.
Swindle, E. J., J. E. Collins, et al. (2009). "Breakdown in epithelial barrier function in patients with asthma: identification of novel therapeutic approaches." J Allergy Clin Immunol 124(1): 23-34; quiz 35-26.
Swindle, E. J. and D. E. Davies (2011). "Artificial airways for the study of respiratory disease." Expert Rev Respir Med 5(6): 757-765.
SynairgenResearchLtd. (2012, February 2012). "Comparing Efficacy and Safety of Inhaled SNG001 to Placebo." Retrieved 10/12/2012, 2012, from http://clinicaltrials.gov/ct2/show/NCT01126177?term=SG005&rank=1.
Sze, M. A., P. A. Dimitriu, et al. (2012). "The Lung Tissue Microbiome in Chronic Obstructive Pulmonary Disease." Am J Respir Crit Care Med.
Takahashi, T., Y. Kuniyasu, et al. (1998). "Immunologic self-tolerance maintained by CD25+CD4+ naturally anergic and suppressive T cells: induction of autoimmune disease by breaking their anergic/suppressive state." Int Immunol 10(12): 1969-1980.
Takeda, K., S. W. Dow, et al. (2009). "Vaccine-induced CD8+ T cell-dependent suppression of airway hyperresponsiveness and inflammation." J Immunol 183(1): 181-190.
Talbot, T. R., T. V. Hartert, et al. (2005). "Asthma as a risk factor for invasive pneumococcal disease." N Engl J Med 352(20): 2082-2090.
Taniguchi, M., H. Koseki, et al. (1996). "Essential requirement of an invariant V alpha 14 T cell antigen receptor expression in the development of natural killer T cells." Proc Natl Acad Sci U S A 93(20): 11025-11028.
Taylor, D. R., M. W. Pijnenburg, et al. (2006). "Exhaled nitric oxide measurements: clinical application and interpretation." Thorax 61(9): 817-827.
ten Brinke, A., J. T. van Dissel, et al. (2001). "Persistent airflow limitation in adult-onset nonatopic asthma is associated with serologic evidence of Chlamydia pneumoniae infection." J Allergy Clin Immunol 107(3): 449-454.
Thavagnanam, S., J. Fleming, et al. (2008). "A meta-analysis of the association between Caesarean section and childhood asthma." Clin Exp Allergy 38(4): 629-633.
ThermoScientific (2008). NanoDrop 1000 Spectrophotometer. Thomas, S. Y., Y. H. Chyung, et al. (2010). "Natural killer T cells are not the predominant T cell in
asthma and likely modulate, not cause, asthma." J Allergy Clin Immunol 125(5): 980-984. Thornton, A. M. and E. M. Shevach (1998). "CD4+CD25+ immunoregulatory T cells suppress
polyclonal T cell activation in vitro by inhibiting interleukin 2 production." J Exp Med 188(2): 287-296.
Till, S., B. Li, et al. (1995). "Secretion of the eosinophil-active cytokines interleukin-5, granulocyte/macrophage colony-stimulating factor and interleukin-3 by bronchoalveolar lavage CD4+ and CD8+ T cell lines in atopic asthmatics, and atopic and non-atopic controls." Eur J Immunol 25(10): 2727-2731.
Till, S. J., S. R. Durham, et al. (1998). "Allergen-induced proliferation and interleukin-5 production by bronchoalveolar lavage and blood T cells after segmental allergen challenge." Am J Respir Crit Care Med 158(2): 404-411.
Tilloy, F., J. P. Di Santo, et al. (1999). "Thymic dependence of invariant V alpha 14+ natural killer-T cell development." Eur J Immunol 29(10): 3313-3318.
Tilloy, F., E. Treiner, et al. (1999). "An invariant T cell receptor alpha chain defines a novel TAP-independent major histocompatibility complex class Ib-restricted alpha/beta T cell subpopulation in mammals." J Exp Med 189(12): 1907-1921.
Torchinsky, M. B., J. Garaude, et al. (2009). "Innate immune recognition of infected apoptotic cells directs T(H)17 cell differentiation." Nature 458(7234): 78-82.
Treiner, E., L. Duban, et al. (2003). "Selection of evolutionarily conserved mucosal-associated invariant T cells by MR1." Nature 422(6928): 164-169.
Treiner, E., L. Duban, et al. (2005). "Mucosal-associated invariant T (MAIT) cells: an evolutionarily conserved T cell subset." Microbes Infect 7(3): 552-559.
Treiner, E. and O. Lantz (2006). "CD1d- and MR1-restricted invariant T cells: of mice and men." Curr Opin Immunol 18(5): 519-526.
Trompette, A., S. Divanovic, et al. (2009). "Allergenicity resulting from functional mimicry of a Toll-like receptor complex protein." Nature 457(7229): 585-588.
Timothy SC Hinks References
251
Turtle, C. J., H. M. Swanson, et al. (2009). "A distinct subset of self-renewing human memory CD8+ T cells survives cytotoxic chemotherapy." Immunity 31(5): 834-844.
Uddin, M., G. Nong, et al. (2010). "Prosurvival activity for airway neutrophils in severe asthma." Thorax 65(8): 684-689.
Umemura, M., A. Yahagi, et al. (2007). "IL-17-mediated regulation of innate and acquired immune response against pulmonary Mycobacterium bovis bacille Calmette-Guerin infection." J Immunol 178(6): 3786-3796.
Umetsu, D. T. and R. H. Dekruyff (2006). "Immune dysregulation in asthma." Curr Opin Immunol 18(6): 727-732.
Upton, M. N., A. McConnachie, et al. (2000). "Intergenerational 20 year trends in the prevalence of asthma and hay fever in adults: the Midspan family study surveys of parents and offspring." BMJ 321(7253): 88-92.
Urbanski, G., P. Rivereau, et al. (2012). "Whipple disease revealed by lung involvement: a case report and literature review." Chest 141(6): 1595-1598.
Vaccher, S., R. Cordiali, et al. (2007). "Bacteremia caused by Rothia mucilaginosa in a patient with Shwachman-Diamond syndrome." Infection 35(3): 209-210.
Vael, C., L. Vanheirstraeten, et al. (2011). "Denaturing gradient gel electrophoresis of neonatal intestinal microbiota in relation to the development of asthma." BMC Microbiol 11: 68.
van Boxel-Dezaire, A. H., J. A. Zula, et al. (2010). "Major differences in the responses of primary human leukocyte subsets to IFN-beta." J Immunol 185(10): 5888-5899.
van Rensen, E. L., J. K. Sont, et al. (2005). "Bronchial CD8 cell infiltrate and lung function decline in asthma." Am J Respir Crit Care Med 172(7): 837-841.
Vanaudenaerde, B. M., S. E. Verleden, et al. (2011). "Innate and Adaptive Interleukin-17-producing Lymphocytes in Chronic Inflammatory Lung Disorders." Am J Respir Crit Care Med 183(8): 977-986.
Vazquez-Tello, A., R. Halwani, et al. (2012). "IL-17A and IL-17F expression in B lymphocytes." Int Arch Allergy Immunol 157(4): 406-416.
Vazquez-Tello, A., A. Semlali, et al. (2010). "Induction of glucocorticoid receptor-beta expression in epithelial cells of asthmatic airways by T-helper type 17 cytokines." Clin Exp Allergy 40(9): 1312-1322.
Veldhoen, M. and B. Stockinger (2006). "TGFbeta1, a "Jack of all trades": the link with pro-inflammatory IL-17-producing T cells." Trends Immunol 27(8): 358-361.
Vijayanand, P. (2007). A translational study of the mechanisms of accumulation and activation of T cells in human asthmatic airway. Doctor of Philosophy, University of Southampton.
Vijayanand, P., G. Seumois, et al. (2007). "Invariant natural killer T cells in asthma and chronic obstructive pulmonary disease." N Engl J Med 356(14): 1410-1422.
Vijayanand, P., G. Seumois, et al. (2012). "Interleukin-4 production by follicular helper T cells requires the conserved Il4 enhancer hypersensitivity site V." Immunity 36(2): 175-187.
Virgin, H. W., E. J. Wherry, et al. (2009). "Redefining chronic viral infection." Cell 138(1): 30-50. Voehringer, D. (2011). "Basophils in allergic immune responses." Curr Opin Immunol 23(6): 789-793. Voelkerding, K. V., S. A. Dames, et al. (2009). "Next-generation sequencing: from basic research to
diagnostics." Clin Chem 55(4): 641-658. von Mutius, E. (2007). "Allergies, infections and the hygiene hypothesis--the epidemiological
evidence." Immunobiology 212(6): 433-439. von Mutius, E., C. Fritzsch, et al. (1992). "Prevalence of asthma and allergic disorders among children
in united Germany: a descriptive comparison." BMJ 305(6866): 1395-1399. Von Pirquet, C. (1906). "Allergie." München. Med. Wchnschr. 53: 1457. von Pirquet, C. and B. Schick (1905). die Serumkrankhit. Wien, Deuticke, F. Voo, K. S., Y. H. Wang, et al. (2009). "Identification of IL-17-producing FOXP3+ regulatory T cells in
humans." Proc Natl Acad Sci U S A 106(12): 4793-4798. Wakashin, H., K. Hirose, et al. (2008). "IL-23 and Th17 Cells Enhance Th2 Cell-mediated Eosinophilic
Airway Inflammation in Mice." Am J Respir Crit Care Med. Walker, C., W. Bauer, et al. (1994). "Activated T cells and cytokines in bronchoalveolar lavages from
patients with various lung diseases associated with eosinophilia." Am J Respir Crit Care Med 150(4): 1038-1048.
Walker, C., M. K. Kaegi, et al. (1991). "Activated T cells and eosinophilia in bronchoalveolar lavages from subjects with asthma correlated with disease severity." J Allergy Clin Immunol 88(6): 935-942.
Walker, L. J., Y. H. Kang, et al. (2012). "Human MAIT and CD8alphaalpha cells develop from a pool of type-17 precommitted CD8+ T cells." Blood 119(2): 422-433.
Timothy SC Hinks References
252
Walter, M. J. and M. J. Holtzman (2005). "A centennial history of research on asthma pathogenesis." Am J Respir Cell Mol Biol 32(6): 483-489.
Wang, D., H. Zhang, et al. "CD4+ CD25+ but not CD4+ Foxp3+ T cells as a regulatory subset in primary biliary cirrhosis." Cell Mol Immunol 7(6): 485-490.
Wang, L. H., Y. H. Lin, et al. (2009). "Insufficient increment of CD4+CD25+ regulatory T cells after stimulation in vitro with allergen in allergic asthma." Int Arch Allergy Immunol 148(3): 199-210.
Wang, Y. H., K. S. Voo, et al. (2010). "A novel subset of CD4(+) T(H)2 memory/effector cells that produce inflammatory IL-17 cytokine and promote the exacerbation of chronic allergic asthma." J Exp Med 207(11): 2479-2491.
Wardlaw, A. J., C. Brightling, et al. (2000). "Eosinophils in asthma and other allergic diseases." Br Med Bull 56(4): 985-1003.
Wark, P. A., S. L. Johnston, et al. (2005). "Asthmatic bronchial epithelial cells have a deficient innate immune response to infection with rhinovirus." J Exp Med 201(6): 937-947.
Wark, P. A., S. L. Johnston, et al. (2002). "Neutrophil degranulation and cell lysis is associated with clinical severity in virus-induced asthma." Eur Respir J 19(1): 68-75.
Weeks-Gorospe, J. N., H. R. Hurtig, et al. (2012). "Naturally occurring swine influenza A virus PB1-F2 phenotypes that contribute to superinfection with Gram-positive respiratory pathogens." J Virol 86(17): 9035-9043.
Weichsel, R., C. Dix, et al. (2008). "Profound inhibition of antigen-specific T-cell effector functions by dasatinib." Clin Cancer Res 14(8): 2484-2491.
Weinberg, A., L. Y. Song, et al. (2009). "Optimization and limitations of use of cryopreserved peripheral blood mononuclear cells for functional and phenotypic T-cell characterization." Clin Vaccine Immunol 16(8): 1176-1186.
Weinberg, A., D. A. Wohl, et al. (2000). "Effect of cryopreservation on measurement of cytomegalovirus-specific cellular immune responses in HIV-infected patients." J Acquir Immune Defic Syndr 25(2): 109-114.
Weiss, S. T. and A. A. Litonjua (2011). "Vitamin D in asthma and allergy: what next?" Eur Respir J 38(6): 1255-1257.
Welsh, E. J., C. J. Cates, et al. (2010). "Combination inhaled steroid and long-acting beta2-agonist versus tiotropium for chronic obstructive pulmonary disease." Cochrane Database Syst Rev(5): CD007891.
Wenink, M. H., K. C. Santegoets, et al. (2009). "TLR2 promotes Th2/Th17 responses via TLR4 and TLR7/8 by abrogating the type I IFN amplification loop." J Immunol 183(11): 6960-6970.
Wenzel, S. E., S. J. Szefler, et al. (1997). "Bronchoscopic evaluation of severe asthma. Persistent inflammation associated with high dose glucocorticoids." Am J Respir Crit Care Med 156(3 Pt 1): 737-743.
Wickens, K., N. Pearce, et al. (1999). "Antibiotic use in early childhood and the development of asthma." Clin Exp Allergy 29(6): 766-771.
Wiehler, S. and D. Proud (2007). "Interleukin-17A modulates human airway epithelial responses to human rhinovirus infection." Am J Physiol Lung Cell Mol Physiol 293(2): L505-515.
Wilkinson, T. M., I. S. Patel, et al. (2003). "Airway bacterial load and FEV1 decline in patients with chronic obstructive pulmonary disease." Am J Respir Crit Care Med 167(8): 1090-1095.
Willing, B. P., J. Dicksved, et al. (2010). "A pyrosequencing study in twins shows that gastrointestinal microbial profiles vary with inflammatory bowel disease phenotypes." Gastroenterology 139(6): 1844-1854 e1841.
Wills-Karp, M., D. Brandt, et al. (2004). "Understanding the origin of asthma and its relationship to breastfeeding." Adv Exp Med Biol 554: 171-191.
Wilson, R. H., G. S. Whitehead, et al. (2009). "Allergic sensitization through the airway primes Th17-dependent neutrophilia and airway hyperresponsiveness." Am J Respir Crit Care Med 180(8): 720-730.
Wimalasundera, S. S., D. R. Katz, et al. (1997). "Characterization of the T cell response to human rhinovirus in children: implications for understanding the immunopathology of the common cold." J Infect Dis 176(3): 755-759.
Wingender, G., P. Rogers, et al. (2011). "Invariant NKT cells are required for airway inflammation induced by environmental antigens." J Exp Med 208(6): 1151-1162.
Woellner, C., E. M. Gertz, et al. (2010). "Mutations in STAT3 and diagnostic guidelines for hyper-IgE syndrome." J Allergy Clin Immunol 125(2): 424-432 e428.
Wolk, K., S. Kunz, et al. (2004). "IL-22 increases the innate immunity of tissues." Immunity 21(2): 241-254.
Timothy SC Hinks References
253
Wong, C. K., S. W. Lun, et al. (2009). "Activation of peripheral Th17 lymphocytes in patients with asthma." Immunol Invest 38(7): 652-664.
Woodruff, P. G., B. Modrek, et al. (2009). "T-helper type 2-driven inflammation defines major subphenotypes of asthma." Am J Respir Crit Care Med 180(5): 388-395.
Wos, M., M. Sanak, et al. (2008). "The presence of rhinovirus in lower airways of patients with bronchial asthma." Am J Respir Crit Care Med 177(10): 1082-1089.
Wu, A. C., K. Tantisira, et al. (2012). "Effect of vitamin D and inhaled corticosteroid treatment on lung function in children." Am J Respir Crit Care Med 186(6): 508-513.
Wu, K., Y. Bi, et al. (2008). "Suppression of allergic inflammation by allergen-DNA-modified dendritic cells depends on the induction of Foxp3+ Regulatory T cells." Scand J Immunol 67(2): 140-151.
Xu, G., L. Zhang, et al. (2010). "Opposing roles of IL-17A and IL-25 in the regulation of TSLP production in human nasal epithelial cells." Allergy 65(5): 581-589.
Yang, L., D. E. Anderson, et al. (2008). "IL-21 and TGF-beta are required for differentiation of human T(H)17 cells." Nature 454(7202): 350-352.
Yao, Z., W. C. Fanslow, et al. (1995). "Herpesvirus Saimiri encodes a new cytokine, IL-17, which binds to a novel cytokine receptor." Immunity 3(6): 811-821.
Ying, S., M. Humbert, et al. (1997). "Expression of IL-4 and IL-5 mRNA and protein product by CD4+ and CD8+ T cells, eosinophils, and mast cells in bronchial biopsies obtained from atopic and nonatopic (intrinsic) asthmatics." J Immunol 158(7): 3539-3544.
Yokote, H., S. Miyake, et al. (2008). "NKT cell-dependent amelioration of a mouse model of multiple sclerosis by altering gut flora." Am J Pathol 173(6): 1714-1723.
Yoshida, M., R. M. Watson, et al. (2005). "Different profiles of T-cell IFN-gamma and IL-12 in allergen-induced early and dual responders with asthma." J Allergy Clin Immunol 115(5): 1004-1009.
Yu, S. and M. T. Cantorna (2011). "Epigenetic reduction in invariant NKT cells following in utero vitamin D deficiency in mice." J Immunol 186(3): 1384-1390.
Yu, S., J. Zhao, et al. (2011). "Invariant NKT cell defects in vitamin D receptor knockout mice prevents experimental lung inflammation." J Immunol 187(9): 4907-4912.
Zaura, E., B. J. Keijser, et al. (2009). "Defining the healthy "core microbiome" of oral microbial communities." BMC Microbiol 9: 259.
Zhang, X., J. Jin, et al. (2009). "IFN-beta1a inhibits the secretion of Th17-polarizing cytokines in human dendritic cells via TLR7 up-regulation." J Immunol 182(6): 3928-3936.
Zhang, Y., M. F. Moffatt, et al. (2012). "Genetic and genomic approaches to asthma: new insights for the origins." Curr Opin Pulm Med 18(1): 6-13.
Zhao, G., Cai, Z., Cao, S., Virgin, H.W., Wang, D. VirusHunter : An Automated Data Analysis Pipeline For Virus Discovery
Zhao, Y., J. Yang, et al. (2010). "Th17 immunity in patients with allergic asthma." Int Arch Allergy Immunol 151(4): 297-307.
Zhou, L., J. E. Lopes, et al. (2008). "TGF-beta-induced Foxp3 inhibits T(H)17 cell differentiation by antagonizing RORgammat function." Nature 453(7192): 236-240.
Zhou, Q. T., Y. C. Sun, et al. (2005). "[Characteristics of the airway inflammation and the relationship to interleukin-17 in severe asthma]." Zhonghua Jie He He Hu Xi Za Zhi 28(9): 630-634.
Zosky, G. R. and P. D. Sly (2007). "Animal models of asthma." Clin Exp Allergy 37(7): 973-988. Zuany-Amorim, C., C. Ruffie, et al. (1998). "Requirement for gammadelta T cells in allergic airway
inflammation." Science 280(5367): 1265-1267. Zuany-Amorim, C., E. Sawicka, et al. (2002). "Suppression of airway eosinophilia by killed
Mycobacterium vaccae-induced allergen-specific regulatory T-cells." Nat Med 8(6): 625-629.